Report.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Introduction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Do not spend too much time trying to get very tiny metrics improvement. Once you have a model with a correct predictive power, you should better spend time explaining your data cleaning & preparation pipeline as well as explanations & visualizations of the results.\n",
    "\n",
    "The goal is to see your fit with our company culture & engineering needs, spending 50h on an over-complicated approach will not give you bonus points compared to a simple, yet effective, to-the-point solution."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## About the data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The dataset you will be working with is called Emo-DB and can be found [here](http://emodb.bilderbar.info/index-1280.html).\n",
    "\n",
    "It is a database containing samples of emotional speech in German. It contains samples labeled with one of 7 different emotions: Anger, Boredom, Disgust, Fear, Happiness, Sadness and Neutral. \n",
    "\n",
    "Please download the full database and refer to the documentation to understand how the samples are labeled (see \"Additional information\")\n",
    "   \n",
    "The goal of this project is to develop a model which is able to **classify samples of emotional speech**. Feel free to use any available library you would need, but beware of re-using someone else's code without mentionning it!"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Deliverable"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The end-goal is to deliver us:\n",
    "* This report filled with your approach, in the form of an **iPython Notebook**.\n",
    "* A **5-10 slides PDF file**, containing a very brief presentation covering the following points:\n",
    "    * Introduction to the problem (what are we trying to achieve and why) - max 1 slide\n",
    "    * Libraries used - max 1 slide\n",
    "    * Data Processing Pipeline - max 2 slides\n",
    "    * Feature Engineering (if relevant) - max 1 slide\n",
    "    * Modeling - max 1 slide\n",
    "    * Results & Visualization - max 2 slides\n",
    "* The goal of the presentation is to make it **understandable by a business person**, apart from how modeling techniques which do not have to be explained how they work."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Libraries Loading"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import platform\n",
    "import glob\n",
    "import ntpath\n",
    "import os\n",
    "from scipy.io import wavfile"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "import torch\n",
    "#from torch import Tensor\n",
    "from torch import nn\n",
    "import torch.nn.functional as F"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/shernand/dl_epfl/lib/python3.6/site-packages/pandas_profiling/plot.py:15: UserWarning: \n",
      "This call to matplotlib.use() has no effect because the backend has already\n",
      "been chosen; matplotlib.use() must be called *before* pylab, matplotlib.pyplot,\n",
      "or matplotlib.backends is imported for the first time.\n",
      "\n",
      "The backend was *originally* set to 'module://ipykernel.pylab.backend_inline' by the following code:\n",
      "  File \"/home/shernand/pythonroots/python_3.6.5_built/lib/python3.6/runpy.py\", line 193, in _run_module_as_main\n",
      "    \"__main__\", mod_spec)\n",
      "  File \"/home/shernand/pythonroots/python_3.6.5_built/lib/python3.6/runpy.py\", line 85, in _run_code\n",
      "    exec(code, run_globals)\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/ipykernel_launcher.py\", line 16, in <module>\n",
      "    app.launch_new_instance()\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/traitlets/config/application.py\", line 658, in launch_instance\n",
      "    app.start()\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/ipykernel/kernelapp.py\", line 486, in start\n",
      "    self.io_loop.start()\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/tornado/platform/asyncio.py\", line 127, in start\n",
      "    self.asyncio_loop.run_forever()\n",
      "  File \"/home/shernand/pythonroots/python_3.6.5_built/lib/python3.6/asyncio/base_events.py\", line 422, in run_forever\n",
      "    self._run_once()\n",
      "  File \"/home/shernand/pythonroots/python_3.6.5_built/lib/python3.6/asyncio/base_events.py\", line 1432, in _run_once\n",
      "    handle._run()\n",
      "  File \"/home/shernand/pythonroots/python_3.6.5_built/lib/python3.6/asyncio/events.py\", line 145, in _run\n",
      "    self._callback(*self._args)\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/tornado/platform/asyncio.py\", line 117, in _handle_events\n",
      "    handler_func(fileobj, events)\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/tornado/stack_context.py\", line 276, in null_wrapper\n",
      "    return fn(*args, **kwargs)\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/zmq/eventloop/zmqstream.py\", line 450, in _handle_events\n",
      "    self._handle_recv()\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/zmq/eventloop/zmqstream.py\", line 480, in _handle_recv\n",
      "    self._run_callback(callback, msg)\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/zmq/eventloop/zmqstream.py\", line 432, in _run_callback\n",
      "    callback(*args, **kwargs)\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/tornado/stack_context.py\", line 276, in null_wrapper\n",
      "    return fn(*args, **kwargs)\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/ipykernel/kernelbase.py\", line 283, in dispatcher\n",
      "    return self.dispatch_shell(stream, msg)\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/ipykernel/kernelbase.py\", line 233, in dispatch_shell\n",
      "    handler(stream, idents, msg)\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/ipykernel/kernelbase.py\", line 399, in execute_request\n",
      "    user_expressions, allow_stdin)\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/ipykernel/ipkernel.py\", line 208, in do_execute\n",
      "    res = shell.run_cell(code, store_history=store_history, silent=silent)\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/ipykernel/zmqshell.py\", line 537, in run_cell\n",
      "    return super(ZMQInteractiveShell, self).run_cell(*args, **kwargs)\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/IPython/core/interactiveshell.py\", line 2666, in run_cell\n",
      "    self.events.trigger('post_run_cell', result)\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/IPython/core/events.py\", line 88, in trigger\n",
      "    func(*args, **kwargs)\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/ipykernel/pylab/backend_inline.py\", line 160, in configure_once\n",
      "    activate_matplotlib(backend)\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/IPython/core/pylabtools.py\", line 311, in activate_matplotlib\n",
      "    matplotlib.pyplot.switch_backend(backend)\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/matplotlib/pyplot.py\", line 231, in switch_backend\n",
      "    matplotlib.use(newbackend, warn=False, force=True)\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/matplotlib/__init__.py\", line 1410, in use\n",
      "    reload(sys.modules['matplotlib.backends'])\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/importlib/__init__.py\", line 166, in reload\n",
      "    _bootstrap._exec(spec, module)\n",
      "  File \"/home/shernand/dl_epfl/lib/python3.6/site-packages/matplotlib/backends/__init__.py\", line 16, in <module>\n",
      "    line for line in traceback.format_stack()\n",
      "\n",
      "\n",
      "  matplotlib.use(BACKEND)\n"
     ]
    }
   ],
   "source": [
    "import pandas_profiling\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import json"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Preparation & Cleaning"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_main_folder = 'data'\n",
    "wav_folder = 'wav'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "platform_system = platform.system()\n",
    "file_separator=''\n",
    "if(platform_system=='Windows'):\n",
    "    file_separator = '\\\\'\n",
    "else:#(platform_system=='Linux'):\n",
    "    file_separator = '/'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "wav_path = data_main_folder + file_separator + wav_folder"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "wav_files = glob.glob(wav_path+file_separator+'*.wav')\n",
    "wav_files.sort()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['data/wav/03a01Fa.wav']"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "wav_files[:1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## About the data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h2>Additional Information</h2>\n",
    "\n",
    "\n",
    "Every utterance is named according to the same scheme:\n",
    "<ul><li>Positions 1-2: number of speaker\n",
    "    <li>Positions 3-5: code for text\n",
    "    <li>Position 6: emotion (sorry, letter stands for german emotion word)\n",
    "    <li>Position 7: if there are more than two versions these are numbered a, b, c ....\n",
    "</ul>\n",
    "\n",
    "Example: 03a01Fa.wav is the audio file from Speaker 03 speaking text a01 with the emotion \"Freude\" (Happiness).\n",
    "\n",
    "<h3>Information about the speakers</h3>\n",
    "\n",
    "<ul><li>03 - male, 31 years old\n",
    "    <li>08 - female, 34 years\n",
    "    <li>09 - female, 21 years\n",
    "    <li>10 - male, 32 years\n",
    "    <li>11 - male, 26 years\n",
    "    <li>12 - male, 30 years\n",
    "    <li>13 - female, 32 years\n",
    "    <li>14 - female, 35 years\n",
    "    <li>15 - male, 25 years\n",
    "    <li>16 - female, 31 years\n",
    "</ul>\n",
    "\n",
    "<table class=\"aussen\" border=\"0\" align=\"left\" width=\"100%\">\n",
    "<tr><td>\n",
    "\n",
    "<h3>Code of texts</h3>\n",
    "\n",
    "<table align=\"left\" width=\"800\" bgcolor=\"#313131\" border=\"1\" frame=\"solid\" rules=\"rows\" cellpadding=\"3\" cellspacing=\"1\">\n",
    "<tr><th>code</th><th>text (german)</th><th>try of an english translation</th></tr>\n",
    "<tr><td class=\"mittig\">a01</td><td>Der Lappen liegt auf dem Eisschrank.</td><td>The tablecloth is lying on the frigde.</td></tr>\n",
    "<tr><td class=\"mittig\">a02</td><td>Das will sie am Mittwoch abgeben.</td><td>She will hand it in on Wednesday.</td></tr>\n",
    "<tr><td class=\"mittig\">a04</td><td>Heute abend könnte ich es ihm sagen.</td><td>Tonight I could tell him.</td></tr>\n",
    "<tr><td class=\"mittig\">a05</td><td>Das schwarze Stück Papier befindet sich da oben neben dem Holzstück.</td><td>The black sheet of paper is located up there besides the piece of timber.</td></tr>\n",
    "<tr><td class=\"mittig\">a07</td><td>In sieben Stunden wird es soweit sein.</td><td>In seven hours it will be.</td></tr>\n",
    "<tr><td class=\"mittig\">b01</td><td>Was sind denn das für Tüten, die da unter dem Tisch stehen?</td><td>What about the bags standing there under the table?</td></tr>\n",
    "<tr><td class=\"mittig\">b02</td><td>Sie haben es gerade hochgetragen und jetzt gehen sie wieder runter.</td><td>They just carried it upstairs and now they are going down again.</td></tr>\n",
    "<tr><td class=\"mittig\">b03</td><td>An den Wochenenden bin ich jetzt immer nach Hause gefahren und habe Agnes besucht.</td><td>Currently at the weekends I always went home and saw Agnes.</td></tr>\n",
    "<tr><td class=\"mittig\">b09</td><td>Ich will das eben wegbringen und dann mit Karl was trinken gehen.</td><td>I will just discard this and then go for a drink with Karl.</td></tr>\n",
    "<tr><td class=\"mittig\">b10</td><td>Die wird auf dem Platz sein, wo wir sie immer hinlegen.</td><td>It will be in the place where we always store it.</td></tr>\n",
    "</table>\n",
    "<p>&nbsp;<br><p>\n",
    "\n",
    "</td></tr>\n",
    "\n",
    "<tr><td>\n",
    "\n",
    "<h3><br>Code of emotions:</h3>\n",
    "\n",
    "<table align=\"left\" width=\"600\" bgcolor=\"#313131\" border=\"1\" frame=\"solid\" rules=\"rows\" cellpadding=\"3\" cellspacing=\"1\">\n",
    "<tr><th>letter</th><th>emotion (english)</th><th>letter</th><th>emotion (german)</th></tr>\n",
    "<tr><td class=\"mittig\">A</td><td>anger</td><td class=\"mittig\">W</td><td>Ärger (Wut)</td></tr>\n",
    "<tr><td class=\"mittig\">B</td><td>boredom</td><td class=\"mittig\">L</td><td>Langeweile</td></tr>\n",
    "<tr><td class=\"mittig\">D</td><td>disgust</td><td class=\"mittig\">E</td><td>Ekel</td></tr>\n",
    "<tr><td class=\"mittig\">F</td><td>anxiety/fear</td><td class=\"mittig\">A</td><td>Angst</td></tr>\n",
    "<tr><td class=\"mittig\">H</td><td>happiness</td><td class=\"mittig\">F</td><td>Freude</td></tr>\n",
    "<tr><td class=\"mittig\">S</td><td>sadness</td><td class=\"mittig\">T</td><td>Trauer</td></tr>\n",
    "<tr><td colspan=\"4\">N = neutral version</td></tr>\n",
    "\n",
    "</td></tr>\n",
    "</table>\n",
    "\n",
    "</table>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Constructing the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "emotion_german = ['W','L','E','A','F','T','N']\n",
    "emotion_english = ['Anger','Boredom','Disgust','Fear','Happiness','Sadness','Neutral']\n",
    "emotion_numbers = np.arange(0,7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "def loading_wav_file(file_wav, emotion_german,emotion_numbers):\n",
    "    \"\"\"Loads a wav file, and returns its respective y (label) and x (sound signal).\"\"\"\n",
    "    path, filename = os.path.split(file_wav)\n",
    "    filename, file_extension = os.path.splitext(filename)\n",
    "    individual_number = int(filename[0:2])\n",
    "    code_text = filename[2:5]\n",
    "    emotion_label = filename[5]\n",
    "    index_emotion = np.argwhere(np.char.strip(emotion_german)==emotion_label)[0][0]\n",
    "    y = emotion_numbers[index_emotion]\n",
    "    fs, x = wavfile.read(file_wav) \n",
    "    return y, x, fs, code_text, individual_number"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/home/shernand/dl_epfl/lib/python3.6/site-packages/scipy/io/wavfile.py:273: WavFileWarning: Chunk (non-data) not understood, skipping it.\n",
      "  WavFileWarning)\n"
     ]
    }
   ],
   "source": [
    "y=[]\n",
    "x=[]\n",
    "fs_array=[]\n",
    "code_text_array = []\n",
    "ind_numb_array=[]\n",
    "length_array = []\n",
    "for file_wav in wav_files:\n",
    "    y_file, x_file, fs, code_text, individual_number = loading_wav_file(file_wav, emotion_german,emotion_numbers)\n",
    "    y.append(y_file)\n",
    "    x.append(x_file)\n",
    "    length_array.append(len(x_file))\n",
    "    fs_array.append(fs)\n",
    "    code_text_array.append(code_text)\n",
    "    ind_numb_array.append(individual_number)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_complete = pd.DataFrame(columns=['y','x','length','fs','ind_numb','text'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "x = np.array(x)\n",
    "y = np.array(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_complete['y'] = y\n",
    "data_complete['x'] = x\n",
    "data_complete['length'] = length_array\n",
    "data_complete['fs'] = fs_array\n",
    "data_complete['ind_numb'] = ind_numb_array\n",
    "data_complete['text'] = code_text_array"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Profiling of the data-set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<meta charset=\"UTF-8\">\n",
       "\n",
       "<style>\n",
       "\n",
       "        .variablerow {\n",
       "            border: 1px solid #e1e1e8;\n",
       "            border-top: hidden;\n",
       "            padding-top: 2em;\n",
       "            padding-bottom: 2em;\n",
       "            padding-left: 1em;\n",
       "            padding-right: 1em;\n",
       "        }\n",
       "\n",
       "        .headerrow {\n",
       "            border: 1px solid #e1e1e8;\n",
       "            background-color: #f5f5f5;\n",
       "            padding: 2em;\n",
       "        }\n",
       "        .namecol {\n",
       "            margin-top: -1em;\n",
       "            overflow-x: auto;\n",
       "        }\n",
       "\n",
       "        .dl-horizontal dt {\n",
       "            text-align: left;\n",
       "            padding-right: 1em;\n",
       "            white-space: normal;\n",
       "        }\n",
       "\n",
       "        .dl-horizontal dd {\n",
       "            margin-left: 0;\n",
       "        }\n",
       "\n",
       "        .ignore {\n",
       "            opacity: 0.4;\n",
       "        }\n",
       "\n",
       "        .container.pandas-profiling {\n",
       "            max-width:975px;\n",
       "        }\n",
       "\n",
       "        .col-md-12 {\n",
       "            padding-left: 2em;\n",
       "        }\n",
       "\n",
       "        .indent {\n",
       "            margin-left: 1em;\n",
       "        }\n",
       "\n",
       "        .center-img {\n",
       "            margin-left: auto !important;\n",
       "            margin-right: auto !important;\n",
       "            display: block;\n",
       "        }\n",
       "\n",
       "        /* Table example_values */\n",
       "            table.example_values {\n",
       "                border: 0;\n",
       "            }\n",
       "\n",
       "            .example_values th {\n",
       "                border: 0;\n",
       "                padding: 0 ;\n",
       "                color: #555;\n",
       "                font-weight: 600;\n",
       "            }\n",
       "\n",
       "            .example_values tr, .example_values td{\n",
       "                border: 0;\n",
       "                padding: 0;\n",
       "                color: #555;\n",
       "            }\n",
       "\n",
       "        /* STATS */\n",
       "            table.stats {\n",
       "                border: 0;\n",
       "            }\n",
       "\n",
       "            .stats th {\n",
       "                border: 0;\n",
       "                padding: 0 2em 0 0;\n",
       "                color: #555;\n",
       "                font-weight: 600;\n",
       "            }\n",
       "\n",
       "            .stats tr {\n",
       "                border: 0;\n",
       "            }\n",
       "\n",
       "            .stats td{\n",
       "                color: #555;\n",
       "                padding: 1px;\n",
       "                border: 0;\n",
       "            }\n",
       "\n",
       "\n",
       "        /* Sample table */\n",
       "            table.sample {\n",
       "                border: 0;\n",
       "                margin-bottom: 2em;\n",
       "                margin-left:1em;\n",
       "            }\n",
       "            .sample tr {\n",
       "                border:0;\n",
       "            }\n",
       "            .sample td, .sample th{\n",
       "                padding: 0.5em;\n",
       "                white-space: nowrap;\n",
       "                border: none;\n",
       "\n",
       "            }\n",
       "\n",
       "            .sample thead {\n",
       "                border-top: 0;\n",
       "                border-bottom: 2px solid #ddd;\n",
       "            }\n",
       "\n",
       "            .sample td {\n",
       "                width:100%;\n",
       "            }\n",
       "\n",
       "\n",
       "        /* There is no good solution available to make the divs equal height and then center ... */\n",
       "            .histogram {\n",
       "                margin-top: 3em;\n",
       "            }\n",
       "        /* Freq table */\n",
       "\n",
       "            table.freq {\n",
       "                margin-bottom: 2em;\n",
       "                border: 0;\n",
       "            }\n",
       "            table.freq th, table.freq tr, table.freq td {\n",
       "                border: 0;\n",
       "                padding: 0;\n",
       "            }\n",
       "\n",
       "            .freq thead {\n",
       "                font-weight: 600;\n",
       "                white-space: nowrap;\n",
       "                overflow: hidden;\n",
       "                text-overflow: ellipsis;\n",
       "\n",
       "            }\n",
       "\n",
       "            td.fillremaining{\n",
       "                width:auto;\n",
       "                max-width: none;\n",
       "            }\n",
       "\n",
       "            td.number, th.number {\n",
       "                text-align:right ;\n",
       "            }\n",
       "\n",
       "        /* Freq mini */\n",
       "            .freq.mini td{\n",
       "                width: 50%;\n",
       "                padding: 1px;\n",
       "                font-size: 12px;\n",
       "\n",
       "            }\n",
       "            table.freq.mini {\n",
       "                 width:100%;\n",
       "            }\n",
       "            .freq.mini th {\n",
       "                overflow: hidden;\n",
       "                text-overflow: ellipsis;\n",
       "                white-space: nowrap;\n",
       "                max-width: 5em;\n",
       "                font-weight: 400;\n",
       "                text-align:right;\n",
       "                padding-right: 0.5em;\n",
       "            }\n",
       "\n",
       "            .missing {\n",
       "                color: #a94442;\n",
       "            }\n",
       "            .alert, .alert > th, .alert > td {\n",
       "                color: #a94442;\n",
       "            }\n",
       "\n",
       "\n",
       "        /* Bars in tables */\n",
       "            .freq .bar{\n",
       "                float: left;\n",
       "                width: 0;\n",
       "                height: 100%;\n",
       "                line-height: 20px;\n",
       "                color: #fff;\n",
       "                text-align: center;\n",
       "                background-color: #337ab7;\n",
       "                border-radius: 3px;\n",
       "                margin-right: 4px;\n",
       "            }\n",
       "            .other .bar {\n",
       "                background-color: #999;\n",
       "            }\n",
       "            .missing .bar{\n",
       "                background-color: #a94442;\n",
       "            }\n",
       "            .tooltip-inner {\n",
       "                width: 100%;\n",
       "                white-space: nowrap;\n",
       "                text-align:left;\n",
       "            }\n",
       "\n",
       "            .extrapadding{\n",
       "                padding: 2em;\n",
       "            }\n",
       "\n",
       "            .pp-anchor{\n",
       "\n",
       "            }\n",
       "\n",
       "</style>\n",
       "\n",
       "<div class=\"container pandas-profiling\">\n",
       "    <div class=\"row headerrow highlight\">\n",
       "        <h1>Overview</h1>\n",
       "    </div>\n",
       "    <div class=\"row variablerow\">\n",
       "    <div class=\"col-md-6 namecol\">\n",
       "        <p class=\"h4\">Dataset info</p>\n",
       "        <table class=\"stats\" style=\"margin-left: 1em;\">\n",
       "            <tbody>\n",
       "            <tr>\n",
       "                <th>Number of variables</th>\n",
       "                <td>6 </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Number of observations</th>\n",
       "                <td>535 </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Total Missing (%)</th>\n",
       "                <td>0.0% </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Total size in memory</th>\n",
       "                <td>25.2 KiB </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Average record size in memory</th>\n",
       "                <td>48.1 B </td>\n",
       "            </tr>\n",
       "            </tbody>\n",
       "        </table>\n",
       "    </div>\n",
       "    <div class=\"col-md-6 namecol\">\n",
       "        <p class=\"h4\">Variables types</p>\n",
       "        <table class=\"stats\" style=\"margin-left: 1em;\">\n",
       "            <tbody>\n",
       "            <tr>\n",
       "                <th>Numeric</th>\n",
       "                <td>3 </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Categorical</th>\n",
       "                <td>1 </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Boolean</th>\n",
       "                <td>0 </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Date</th>\n",
       "                <td>0 </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Text (Unique)</th>\n",
       "                <td>0 </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Rejected</th>\n",
       "                <td>1 </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Unsupported</th>\n",
       "                <td>1 </td>\n",
       "            </tr>\n",
       "            </tbody>\n",
       "        </table>\n",
       "    </div>\n",
       "    <div class=\"col-md-12\" style=\"padding-left: 1em;\">\n",
       "        \n",
       "        <p class=\"h4\">Warnings</p>\n",
       "        <ul class=\"list-unstyled\"><li><a href=\"#pp_var_fs\"><code>fs</code></a> has constant value 16000 <span class=\"label label-primary\">Rejected</span></li><li><a href=\"#pp_var_x\"><code>x</code></a> is an unsupported type, check if it needs cleaning or further analysis <span class=\"label label-warning\">Warning</span></li><li><a href=\"#pp_var_y\"><code>y</code></a> has 127 / 23.7% zeros <span class=\"label label-info\">Zeros</span></li> </ul>\n",
       "    </div>\n",
       "</div>\n",
       "    <div class=\"row headerrow highlight\">\n",
       "        <h1>Variables</h1>\n",
       "    </div>\n",
       "    <div class=\"row variablerow ignore\">\n",
       "    <div class=\"col-md-3 namecol\">\n",
       "        <p class=\"h4 pp-anchor\" id=\"pp_var_fs\"><s>fs</s><br/>\n",
       "            <small>Constant</small>\n",
       "        </p>\n",
       "    </div><div class=\"col-md-3\">\n",
       "    <p><em>This variable is constant and should be ignored for analysis</em></p>\n",
       "</div>\n",
       "<div class=\"col-md-6\">\n",
       "    <table class=\"stats \">\n",
       "        <tr>\n",
       "            <th>Constant value</th>\n",
       "            <td>16000</td>\n",
       "        </tr>\n",
       "    </table>\n",
       "</div>\n",
       "</div><div class=\"row variablerow\">\n",
       "    <div class=\"col-md-3 namecol\">\n",
       "        <p class=\"h4 pp-anchor\" id=\"pp_var_ind_numb\">ind_numb<br/>\n",
       "            <small>Numeric</small>\n",
       "        </p>\n",
       "    </div><div class=\"col-md-6\">\n",
       "    <div class=\"row\">\n",
       "        <div class=\"col-sm-6\">\n",
       "            <table class=\"stats \">\n",
       "                <tr>\n",
       "                    <th>Distinct count</th>\n",
       "                    <td>10</td>\n",
       "                </tr>\n",
       "                <tr>\n",
       "                    <th>Unique (%)</th>\n",
       "                    <td>1.9%</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Missing (%)</th>\n",
       "                    <td>0.0%</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Missing (n)</th>\n",
       "                    <td>0</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Infinite (%)</th>\n",
       "                    <td>0.0%</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Infinite (n)</th>\n",
       "                    <td>0</td>\n",
       "                </tr>\n",
       "            </table>\n",
       "\n",
       "        </div>\n",
       "        <div class=\"col-sm-6\">\n",
       "            <table class=\"stats \">\n",
       "\n",
       "                <tr>\n",
       "                    <th>Mean</th>\n",
       "                    <td>11.473</td>\n",
       "                </tr>\n",
       "                <tr>\n",
       "                    <th>Minimum</th>\n",
       "                    <td>3</td>\n",
       "                </tr>\n",
       "                <tr>\n",
       "                    <th>Maximum</th>\n",
       "                    <td>16</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Zeros (%)</th>\n",
       "                    <td>0.0%</td>\n",
       "                </tr>\n",
       "            </table>\n",
       "        </div>\n",
       "    </div>\n",
       "</div>\n",
       "<div class=\"col-md-3 collapse in\" id=\"minihistogram8119680243321151841\">\n",
       "    <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAMgAAABLCAYAAAA1fMjoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD%2BnaQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAAS1JREFUeJzt3MFpw0AARUHbpCQXkZ58dk8pIj2tGwgPS2C0Wc3cBXt5fIEWXccY4wL86Xb0AWBmX0cfgHXdHz%2Bbn/l9fn/gJPtZEAgCgSAQCAKBIBAIAoEgEAgCgSAQCAKBIBAIAoEgEAhu8y5g663Z2W7MzsyCQBAIBIFAEAgEgUAQCASBQPAdhLfs%2BUPJCiwIBAtyQmddgz0sCAQLsoE7T%2BdjQSAIBMJ0r1gr/PCYdVgQCNMtyEqs4f9nQSBYkMn4iDcXCwJBIBAEAkEgEAQCQSAQBAJBIBAEAkEgEAQCQSAQBALhOsYYRx8CZmVBIAgEgkAgCASCQCAIBIJAIAgEgkAgCASCQCAIBIJAIAgEgkAgCASCQCAIBIJAIAgEgkAgCASCQCAIBIJAILwAwTkb0sR3IEUAAAAASUVORK5CYII%3D\">\n",
       "\n",
       "</div>\n",
       "<div class=\"col-md-12 text-right\">\n",
       "    <a role=\"button\" data-toggle=\"collapse\" data-target=\"#descriptives8119680243321151841,#minihistogram8119680243321151841\"\n",
       "       aria-expanded=\"false\" aria-controls=\"collapseExample\">\n",
       "        Toggle details\n",
       "    </a>\n",
       "</div>\n",
       "<div class=\"row collapse col-md-12\" id=\"descriptives8119680243321151841\">\n",
       "    <ul class=\"nav nav-tabs\" role=\"tablist\">\n",
       "        <li role=\"presentation\" class=\"active\"><a href=\"#quantiles8119680243321151841\"\n",
       "                                                  aria-controls=\"quantiles8119680243321151841\" role=\"tab\"\n",
       "                                                  data-toggle=\"tab\">Statistics</a></li>\n",
       "        <li role=\"presentation\"><a href=\"#histogram8119680243321151841\" aria-controls=\"histogram8119680243321151841\"\n",
       "                                   role=\"tab\" data-toggle=\"tab\">Histogram</a></li>\n",
       "        <li role=\"presentation\"><a href=\"#common8119680243321151841\" aria-controls=\"common8119680243321151841\"\n",
       "                                   role=\"tab\" data-toggle=\"tab\">Common Values</a></li>\n",
       "        <li role=\"presentation\"><a href=\"#extreme8119680243321151841\" aria-controls=\"extreme8119680243321151841\"\n",
       "                                   role=\"tab\" data-toggle=\"tab\">Extreme Values</a></li>\n",
       "\n",
       "    </ul>\n",
       "\n",
       "    <div class=\"tab-content\">\n",
       "        <div role=\"tabpanel\" class=\"tab-pane active row\" id=\"quantiles8119680243321151841\">\n",
       "            <div class=\"col-md-4 col-md-offset-1\">\n",
       "                <p class=\"h4\">Quantile statistics</p>\n",
       "                <table class=\"stats indent\">\n",
       "                    <tr>\n",
       "                        <th>Minimum</th>\n",
       "                        <td>3</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>5-th percentile</th>\n",
       "                        <td>3</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Q1</th>\n",
       "                        <td>9</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Median</th>\n",
       "                        <td>12</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Q3</th>\n",
       "                        <td>14</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>95-th percentile</th>\n",
       "                        <td>16</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Maximum</th>\n",
       "                        <td>16</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Range</th>\n",
       "                        <td>13</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Interquartile range</th>\n",
       "                        <td>5</td>\n",
       "                    </tr>\n",
       "                </table>\n",
       "            </div>\n",
       "            <div class=\"col-md-4 col-md-offset-2\">\n",
       "                <p class=\"h4\">Descriptive statistics</p>\n",
       "                <table class=\"stats indent\">\n",
       "                    <tr>\n",
       "                        <th>Standard deviation</th>\n",
       "                        <td>3.7023</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Coef of variation</th>\n",
       "                        <td>0.3227</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Kurtosis</th>\n",
       "                        <td>0.051534</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Mean</th>\n",
       "                        <td>11.473</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>MAD</th>\n",
       "                        <td>3.009</td>\n",
       "                    </tr>\n",
       "                    <tr class=\"\">\n",
       "                        <th>Skewness</th>\n",
       "                        <td>-0.8383</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Sum</th>\n",
       "                        <td>6138</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Variance</th>\n",
       "                        <td>13.707</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Memory size</th>\n",
       "                        <td>4.3 KiB</td>\n",
       "                    </tr>\n",
       "                </table>\n",
       "            </div>\n",
       "        </div>\n",
       "        <div role=\"tabpanel\" class=\"tab-pane col-md-8 col-md-offset-2\" id=\"histogram8119680243321151841\">\n",
       "            <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAlgAAAGQCAYAAAByNR6YAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD%2BnaQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3X9U1HW%2Bx/EXMEki8qsC97KWmmmirj8SuZKJoaXplUwp8p69Lnf7IUViJps/W3Wvq3bFtb3gIa20rE7N4raJqMU107Kos7r9ACK3vHU3OUvMKhPKSgoz949uzE6UonzgO8w8H%2Bd0OHy%2B4/f7mncz%2BPL7nRmC3G63WwAAADAm2OoAAAAA/oaCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMs1kdIFA4HCetjtAhgoODFBPTQydONMjlclsdx3LMw4NZeDALD2bhwSw8OnIWV1zR0%2Bj%2B2oozWGiX4OAgBQUFKTg4yOooPoF5eDALD2bhwSw8mIWHP86CggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhtmsDgAAADrGLY%2B9ZXWENjv068lWRzCKM1gAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMMxvC9abb76p5ORkzZ8/v9W20tJSpaWlacSIEZo0aZJ%2B97vfeW3ftm2bJk2apJEjR2rWrFmqqKjorNgAAMAP2KwO0BGeeOIJbd%2B%2BXVdddVWrbR9%2B%2BKFyc3P1m9/8RuPHj9dbb72l7Oxs9evXT6NGjdK%2BffuUn5%2BvJ598UgMHDtS2bduUlZWl0tJShYWFWXBvAABAV%2BOXZ7BCQ0N/sGA5nU7NmTNHEydOlM1mU0pKigYMGKBDhw5Jkux2u2bMmKFhw4bp0ksv1d133y1Jev311zv1PgAAgK7LL89gzZ49%2Bwe3jRs3TuPGjWv5vqmpSQ6HQ3FxcZKkyspKTZkypWV7cHCwBg0apPLyck2dOrVNx6%2BtrZXD4fBas9nCFBsbeyF3o0sICQn2%2BhromIcHs/BgFh7MwoNZtOZPs/DLgnUh8vLyFBYW1lKqnE6nIiMjvW4TGRmpurq6Nu/TbreroKDAay07O1s5OTntD%2ByjIiK6Wx3BpzAPD2bhwSw8mIUHs/Dwp1kEbMFyu93Ky8tTSUmJtm3bptDQUK9t7ZGRkaHU1FSvNZstTHV1De3ary8KCQlWRER31defVnOzy%2Bo4lmMeHszCg1l4MAsPZtFaR8wiOrqH0f21VUAWLJfLpcWLF%2BvDDz/UCy%2B8oN69e7dsi46OltPp9Lq90%2BnUNddc0%2Bb9x8bGtroc6HCcVFOT/z6Bmptdfn3/LhTz8GAWHszCg1l4MAsPf5qF/1zsvACrV6/WJ5980qpcSdKQIUNUWVnZ8n1zc7M%2B%2BugjDRs2rLNjAgCALirgCtbhw4dVXFyszZs3KyoqqtX2WbNm6eWXX9b777%2Bv06dPq7CwUN26ddP48eM7PywAAOiS/PIS4dChQyV98w5BSdq7d68kqby8XL///e918uRJ3XjjjV5/JjExUVu2bNG4ceP00EMP6cEHH9Tx48c1dOhQbd68WZdeemnn3gkAANBlBbnb%2B4putInDcdLqCB3CZgtWdHQP1dU1%2BM118/ZgHh7MwoNZeDALj86YxS2PvdUh%2B%2B0Ih349uUNmccUVPY3ur60C7hIhAABAR6NgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwvy1Yb775ppKTkzV//vxW23bv3q1p06ZpxIgRmjFjhg4ePNiyzeVyacOGDZowYYISExN111136YsvvujM6AAAoIvzy4L1xBNPaNWqVbrqqqtabauqqtLChQuVm5urd955R5mZmXrggQdUU1MjSXr%2B%2Bee1c%2BdObd68Wa%2B//rr69Omj7Oxsud3uzr4bAACgi/LLghUaGqrt27d/b8EqKipSSkqKUlJSFBoaqrS0NA0YMEDFxcWSJLvdrszMTF199dUKDw/X/PnzdfToUX3wwQedfTcAAEAXZbM6QEeYPXv2D26rrKxUSkqK11pCQoLKy8vV2NioTz/9VAkJCS3bwsPDddVVV6m8vFzDhw9v0/Fra2vlcDi81my2MMXGxl7AvegaQkKCvb4GOubhwSw8mIUHs/BgFq350yz8smCdi9PpVGRkpNdaZGSkPv30U3311Vdyu93fu72urq7Nx7Db7SooKPBay87OVk5OzsUH93EREd2tjuBTmIcHs/BgFh7MwoNZePjTLAKuYEk67%2Bup2vt6q4yMDKWmpnqt2WxhqqtraNd%2BfVFISLAiIrqrvv60mptdVsexHPPwYBYezMKDWXgwi9Y6YhbR0T2M7q%2BtAq5gRUdHy%2Bl0eq05nU7FxMQoKipKwcHB37v9sssua/MxYmNjW10OdDhOqqnJf59Azc0uv75/F4p5eDALD2bhwSw8mIWHP83Cfy52ttGQIUNUUVHhtVZeXq5hw4YpNDRU11xzjSorK1u21dfX6y9/%2BYt%2B8pOfdHZUAADQRQVcwbrjjjv09ttva//%2B/fr666%2B1fft2ff7550pLS5MkzZo1S9u2bdPRo0d16tQp5eXladCgQRo6dKjFyQEAQFfhl5cIvy1DTU1NkqS9e/dK%2BuZM1YABA5SXl6c1a9aourpa/fv316ZNm3TFFVdIku688045HA7927/9mxoaGpSUlNTqBesAAADn4pcFq7y8/Jzbb775Zt18883fuy0oKEg5OTl%2B/Y4/AADQsQLuEiEAAEBHo2ABAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyzWR0AAGDWLY%2B9ZXWENtvz4PVWRwA6BGewAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwLGAL1kcffaTZs2dr1KhRuv7665Wbm6sTJ05IksrKypSenq6RI0dq6tSpKi4utjgtAADoSgKyYDU1Nenee%2B/V8OHD9fbbb6ukpEQnTpzQihUrVFtbq/vvv1933nmnysrKtHTpUj3yyCMqLy%2B3OjYAAOgiArJgORwOORwO3XrrrerWrZuio6N10003qaqqSjt37lSfPn2Unp6u0NBQJScnKzU1VUVFRVbHBgAAXYTN6gDflZqaqhkzZmjmzJn60Y9%2B1CHHiIuL06BBg2S32zVv3jw1NjaqtLRU48ePV2VlpRISErxun5CQoD179rR5/7W1tXI4HF5rNluYYmNjjeT3JSEhwV5fAx3z8GAWHszih9lsgTsTHhet%2BdMsfK5gzZw5U7t27VJhYaHGjBmjO%2B64Q6mpqbLZzEUNDg5Wfn6%2BMjMz9cwzz0iSRo8erQULFuj%2B%2B%2B9XXFyc1%2B2joqJUV1fX5v3b7XYVFBR4rWVnZysnJ6f94X1URER3qyP4FObhwSw8mEVr0dE9rI5gOR4XHv40C58rWNnZ2crOzlZlZaVKSkq0evVqrVy5UtOnT1d6err69u3b7mOcOXNGWVlZmjx5srKysvT3v/9dK1euVG5uroF7IGVkZCg1NdVrzWYLU11dg5H9%2B5KQkGBFRHRXff1pNTe7rI5jOebhwSw8mMUP88efi23F46K1jpiFVSXe5wrWtwYPHqzBgwfr4Ycf1u7du7VixQpt2bJFycnJmjdvnn7yk59c9L7Lysp07NgxPfTQQwoJCVHPnj2Vk5OjW2%2B9VTfccIOcTqfX7evq6hQTE9Pm/cfGxra6HOhwnFRTk/8%2BgZqbXX59/y4U8/BgFh7MojXmwePiH/nTLHz2YufZs2e1e/du3XPPPVq4cKHi4uK0ePFiDRo0SJmZmdq5c%2BdF77u5uVkul0tut7tl7cyZM5Kk5ORkVVRUeN2%2BoqJCw4YNu%2BjjAQCAwOJzZ7COHj2q7du36%2BWXX1ZDQ4MmTZqkZ555Rtddd13LbRITE7VixQpNmzbtoo4xYsQIhYWFKT8/X1lZWWpsbFRhYaESExN16623qqCgQEVFRUpLS9M777yjAwcOyG63m7qLAADAz/lcwZo6dar69u2rOXPmaPr06YqKimp1m5SUlJYPBb0Y0dHReuqpp/Too49q3Lhx6tatm0aPHq0VK1bosssu06ZNm7Rq1SqtXLlS8fHxWrduna699tr23C0AABBAfK5gbdu2TaNHjz7v7T744IN2HWfIkCF69tlnv3dbYmKiduzY0a79AwCAwOVzr8EaOHCgsrKytHfv3pa1p59%2BWvfcc0%2BrF58DAAD4Ip8rWGvWrNHJkyfVv3//lrXx48fL5XJp7dq1FiYDAABoG5%2B7RHjw4EHt3LlT0dHRLWt9%2BvRRXl6e/uVf/sXCZAAAAG3jc2ewGhsbFRoa2mo9ODhYp0%2BftiARAADAhfG5gpWYmKi1a9fqq6%2B%2Baln78ssvtXLlSq%2BPagAAAPBVPneJcMmSJfr5z3%2BuMWPGKDw8XC6XSw0NDerdu/cPvusPAADAl/hcwerdu7d27dqlN954Q3/5y18UHBysvn37auzYsQoJCbE6HgAAwHn5XMGSpG7dumnixIlWxwAAALgoPlewvvjiC61fv16ffPKJGhsbW21/7bXXLEgFAADQdj5XsJYsWaLa2lqNHTtWYWFhVscBAAC4YD5XsCoqKvTaa68pJibG6igAAAAXxec%2BpuGyyy7jzBUAAOjSfK5gzZkzRwUFBXK73VZHAQAAuCg%2Bd4nwjTfe0J/%2B9Ce99NJL%2BvGPf6zgYO8O%2BOKLL1qUDAAAoG18rmCFh4dr3LhxVscAAAC4aD5XsNasWWN1BAAAgHbxuddgSdL//M//KD8/X4sXL25Ze%2B%2B99yxMBAAA0HY%2BV7DKysqUlpam0tJSlZSUSPrmw0dnz57Nh4wCAIAuwecK1oYNG/SLX/xCO3fuVFBQkKRvfj/h2rVrtXHjRovTAQAAnJ/PvQbrz3/%2Bs5577jlJailYkjR58mQtWbLEqlgAAOiWx96yOgK6CJ87g9WzZ8/v/R2EtbW16tatmwWJAAAALozPFayRI0dq9erVOnXqVMvaZ599poULF2rMmDEWJgMAAGgbn7tEuHjxYv3sZz9TUlKSmpubNXLkSJ0%2BfVrXXHON1q5da3U8AACA8/K5gtWrVy%2BVlJTowIED%2Buyzz3TppZeqb9%2B%2Buv76671ekwUAAOCrfK5gSdIll1yiiRMnWh0DAADgovhcwUpNTT3nmSo%2BCwsAAPg6nytYU6ZM8SpYzc3N%2Buyzz1ReXq6f/exnFiYDAABoG58rWLm5ud%2B7/uqrr%2Brdd9/t5DQAAAAXzuc%2BpuGHTJw4Ubt27bI6BgAAwHl1mYL10Ucfye12Wx0DAADgvHzuEuGdd97Zau306dM6evSobr75ZgsSAQAAXBifK1h9%2BvRp9S7C0NBQpaen6/bbb7coFQAAQNv5XMHi09oBAEBX53MF6%2BWXX27zbadPn96BSQAAAC6OzxWspUuXyuVytXpBe1BQkNdaUFAQBQsAAPgknytYTz75pLZs2aKsrCwNHDhQbrdbR44c0RNPPKGf/vSnSkpKsjoiAADAOflcwVq7dq02b96suLi4lrVRo0apd%2B/euuuuu1RSUmJhOgAAgPPzuc/B%2BvzzzxUZGdlqPSIiQtXV1UaPVVhYqLFjx2r48OHKzMzUsWPHJEllZWVKT0/XyJEjNXXqVBUXFxs9LgAA8G8%2BV7Di4%2BO1du1a1dXVtazV19dr/fr1uvLKK40d5/nnn1dxcbG2bdumgwcPqn///nr66adVW1ur%2B%2B%2B/X3feeafKysq0dOlSPfLIIyovLzd2bAAA4N987hLhkiVLtGDBAtntdvXo0UPBwcE6deqULr30Um3cuNHYcbZs2aKFCxeqX79%2BkqRly5ZJkp566in16dNH6enpkqTk5GSlpqaqqKhIQ4cONXZ8AIB0y2NvWR0B6BA%2BV7DGjh2r/fv368CBA6qpqZHb7VZcXJxuuOEG9ezZ08gxvvzySx07dkxfffWVpkyZouPHjyspKUkrVqxQZWWlEhISvG6fkJCgPXv2tHn/tbW1cjgcXms2W5hiY2ON5PclISHBXl8DHfPwYBYezAJoG396jvhcwZKk7t27a8KECaqpqVHv3r2N77%2BmpkaS9Morr2jr1q1yu93KycnRsmXL1NjY6PUCe0mKioryumR5Pna7XQUFBV5r2dnZysnJaX94HxUR0d3qCD6FeXgwCw9mAZybPz1HfK5gNTY2avny5dq1a5ckqaKiQvX19XrooYf0m9/8RhEREe0%2Bxrefp3X33Xe3lKm5c%2BfqnnvuUXJycrv3n5GRodTUVK81my1MdXUN7d63rwkJCVZERHfV159Wc7PL6jiWYx4ezMKDWQBt0xHPkejoHkb311Y%2BV7DWrVunqqoq5eXl6eGHH25Zb25uVl5enn71q1%2B1%2BxiXX365JHmVtfj4eLndbp09e1ZOp9Pr9nV1dYqJiWnz/mNjY1tdDnQ4TqqpyX9/sDY3u/z6/l0o5uHBLDyYBXBu/vQc8bmLna%2B%2B%2Bqr%2B67/%2BS5MnT275pc8RERFas2aNSktLjRyjV69eCg8PV1VVVctadXW1LrnkEqWkpKiiosLr9hUVFRo2bJiRYwMAAP/ncwWroaFBffr0abUeExOjv//970aOYbPZlJ6erscff1z/%2B7//q%2BPHj2vjxo2aNm2abrvtNlVXV6uoqEhff/21Dhw4oAMHDuiOO%2B4wcmwAAOD/fO4S4ZVXXql3331XSUlJXr978JVXXtE//dM/GTvOggULdObMGd1%2B%2B%2B06e/asJk2apGXLlqlHjx7atGmTVq1apZUrVyo%2BPl7r1q3Ttddea%2BzYQEfoSm933/Pg9VZHAIAO5XMF61//9V81d%2B5czZw5Uy6XS1u3blVFRYVeffVVLV261NhxunXrpuXLl2v58uWttiUmJmrHjh3GjgUAAAKLzxWsjIwM2Ww2PffccwoJCdHjjz%2Buvn37Ki8vT5MnT7Y6HgAAwHn5XME6ceKEZs6cqZkzZ1odBQAA4KL43IvcJ0yY4PXaKwAAgK7G5wpWUlLSBf1aGgAAAF/jc5cIf/SjH%2BnXv/61Nm/erCuvvFKXXHKJ1/b169dblAwAAKBtfK5gffrpp%2BrXr58kXdDv/wMAAPAVPlOw5s%2Bfrw0bNujZZ59tWdu4caOys7MtTAUAAHDhfOY1WPv27Wu1tnnzZguSAAAAtI/PFKzve%2Bcg7yYEAABdkc8UrG9/sfP51gAAAHydzxQsAAAAf0HBAgAAMMxn3kV49uxZLViw4LxrfA4WAADwdT5TsK677jrV1taedw0AAMDX%2BUzB%2BsfPvwIAAOjKeA0WAACAYRQsAAAAwyhYAAAAhlGwAAAADPOZF7nj4tzy2FtWR2izPQ9eb3UEAAA6BWewAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhNqsDAAg8tzz2ltURLsieB6%2B3OgKALoYzWAAAAIZRsAAAAAwL%2BIK1evVqDRw4sOX7srIypaena%2BTIkZo6daqKi4stTAcAALqigH4NVlVVlXbs2NHyfW1tre6//34tXbpU06ZN0%2BHDh3Xfffepb9%2B%2BGjp0qIVJAQBAVxKwZ7BcLpeWL1%2BuzMzMlrWdO3eqT58%2BSk9PV2hoqJKTk5WamqqioiLrggIAgC4nYM9gvfjiiwoNDdW0adP02GOPSZIqKyuVkJDgdbuEhATt2bPngvZdW1srh8PhtWazhSk2NrZ9obs4m83/%2B3xISLDXV/iH9j52eVwAbeNPz5GALFh/%2B9vflJ%2Bfr2effdZr3el0Ki4uzmstKipKdXV1F7R/u92ugoICr7Xs7Gzl5ORcXGA/ER3dw%2BoInSYiorvVEWCQqccujwvg3PzpORKQBWvNmjWaMWOG%2Bvfvr2PHjhnff0ZGhlJTU73WbLYw1dU1GD9WVxII9z8kJFgREd1VX39azc0uq%2BPAkPY%2BdnlcAG3TEc8Rq/5xH3AFq6ysTO%2B9955KSkpabYuOjpbT6fRaq6urU0xMzAUdIzY2ttXlQIfjpJqaAvsHayDd/%2BZmV0DdX39n6v8ljwvg3PzpORJwBau4uFjHjx/XjTfeKElyu92SpKSkJP385z9vVbwqKio0bNiwTs8JAAC6roArWIsWLdK8efNavq%2BpqVFGRoZ27Nghl8ulTZs2qaioSGlpaXrnnXd04MAB2e12CxMDAICuJuAKVmRkpCIjI1u%2Bb2pqkiT16tVLkrRp0yatWrVKK1euVHx8vNatW6drr73WkqwAfENX%2B92JAKwXcAXru3784x/ryJEjLd8nJiZ6ffgoAADAhfKfD5wAAADwERQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABgWsAWrurpa2dnZSkpKUnJyshYtWqT6%2BnpJUlVVlX7605/quuuu080336wtW7ZYnBYAAHQlAVuwsrKyFBERoX379umll17SJ598okcffVSNjY2aM2eO/vmf/1lvvvmmNmzYoE2bNqm0tNTqyAAAoIsIyIJVX1%2BvIUOGaMGCBerRo4d69eql2267TYcOHdL%2B/ft19uxZ3XfffQoLC9PgwYN1%2B%2B23y263Wx0bAAB0ETarA1ghIiJCa9as8Vr761//qtjYWFVWVmrgwIEKCQlp2ZaQkKCioqI277%2B2tlYOh8NrzWYLU2xsbPuCd3E2m//3%2BZCQYK%2BvAIC286efnQFZsL6rvLxczz33nAoLC7Vnzx5FRER4bY%2BKipLT6ZTL5VJw8Pn/59vtdhUUFHitZWdnKycnx2juriY6uofVETpNRER3qyMAQJfjTz87A75gHT58WPfdd58WLFig5ORk7dmz53tvFxQU1OZ9ZmRkKDU11WvNZgtTXV1Du7J2dYFw/0NCghUR0V319afV3OyyOg4AdCkd8bPTqn/cB3TB2rdvn37xi1/okUce0fTp0yVJMTEx%2Bvzzz71u53Q6FRUV1aazV5IUGxvb6nKgw3FSTU2B/RduIN3/5mZXQN1fADDBn352%2Bs/Fzgv0pz/9SQsXLtRvf/vblnIlSUOGDNGRI0fU1NTUslZeXq5hw4ZZERMAAHRBAVmwmpqatGzZMuXm5mrs2LFe21JSUhQeHq7CwkKdPn1aH3zwgbZv365Zs2ZZlBYAAHQ1AVmw3n//fR09elSrVq3S0KFDvf5zOBx6/PHH9fbbb2v06NF68MEHNX/%2BfI0fP97q2AAAoIsIyNdgjRo1SkeOHDnnbV544YVOSgMAAPxNQJ7BAgAA6EgULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsH6HtXV1br33nuVlJSkG2%2B8UevWrZPL5bI6FgAA6CJsVgfwRXPnztXgwYO1d%2B9eHT9%2BXHPmzNHll1%2Buf//3f7c6GgAA6AI4g/Ud5eXl%2Bvjjj5Wbm6uePXuqT58%2ByszMlN1utzoaAADoIjiD9R2VlZWKj49XZGRky9rgwYP12Wef6dSpUwoPDz/vPmpra%2BVwOLzWbLYwxcbGGs/bldhs/t/nQ0KCvb4CANrOn352UrC%2Bw%2Bl0KiIiwmvt27JVV1fXpoJlt9tVUFDgtfbAAw9o7ty55oL%2Bv0O/nmx8nxeitrZWdrtdGRkZAV8gpW/m8cwzT1oyD6sfC9/FY8ODWXgwCw9m4VFbW6v8/Hy/moX/VEWD3G53u/58RkaGXnrpJa//MjIyDKXzLQ6HQwUFBa3O2AUq5uHBLDyYhQez8GAWHv44C85gfUdMTIycTqfXmtPpVFBQkGJiYtq0j9jYWL9p4AAA4MJxBus7hgwZor/%2B9a86ceJEy1p5ebn69%2B%2BvHj16WJguyGcEAAAHrklEQVQMAAB0FRSs70hISNDQoUO1fv16nTp1SkePHtXWrVs1a9Ysq6MBAIAuImTFihUrrA7ha2644QaVlJToP/7jP7Rr1y6lp6frrrvuUlBQkNXRfFKPHj00evRozvD9P%2BbhwSw8mIUHs/BgFh7%2BNosgd3tf0Q0AAAAvXCIEAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBSNWr16tgQMHWh3DUoWFhRo7dqyGDx%2BuzMxMHTt2zOpIlvjoo480e/ZsjRo1Stdff71yc3O9fnm6v3vzzTeVnJys%2BfPnt9q2e/duTZs2TSNGjNCMGTN08OBBCxJ2nnPNorS0VGlpaRoxYoQmTZqk3/3udxYk7DznmsW3GhoaNH78eC1atKgTk3W%2Bc83iyy%2B/1H333afhw4crOTlZ69evl8vlsiBl%2B1Gw0G5VVVXasWOH1TEs9fzzz6u4uFjbtm3TwYMH1b9/fz399NNWx%2Bp0TU1NuvfeezV8%2BHC9/fbbKikp0YkTJxQov/L0iSee0KpVq3TVVVe12lZVVaWFCxcqNzdX77zzjjIzM/XAAw%2BopqbGgqQd71yz%2BPDDD5Wbm6ucnBz98Y9/1JIlS/SrX/1Khw4dsiBpxzvXLP5Rfn6%2BTp061UmprHGuWbjdbj3wwAOKj4/XwYMH9eyzz6qsrEzvvvuuBUnbj4KFdnG5XFq%2BfLkyMzOtjmKpLVu2aP78%2BerXr5/Cw8O1bNkyLVu2zOpYnc7hcMjhcOjWW29Vt27dFB0drZtuuklVVVVWR%2BsUoaGh2r59%2B/f%2B5VFUVKSUlBSlpKQoNDRUaWlpGjBggIqLiy1I2vHONQun06k5c%2BZo4sSJstlsSklJ0YABA/y2YJ1rFt/6%2BOOPVVJSottuu60Tk3W%2Bc83ij3/8o7744gs9/PDDCg8P19VXX63t27drzJgxFiRtPwoW2uXFF19UaGiopk2bZnUUy3z55Zc6duyYvvrqK02ZMkVJSUnKyckJqMti34qLi9OgQYNkt9vV0NCg48ePq7S0VOPHj7c6WqeYPXu2evbs%2Bb3bKisrlZCQ4LWWkJCg8vLyzojW6c41i3Hjxik7O7vl%2B6amJjkcDsXFxXVWvE51rllI35y5WbFihebPn6%2BIiIhOTNb5zjWLw4cPa8CAAdqwYYOSkpI0YcIEbdmypZMTmkPBwkX729/%2Bpvz8fC1fvtzqKJb69hLPK6%2B8oq1bt2rHjh2qqakJyDNYwcHBys/P12uvvaaRI0cqOTlZTU1NWrBggdXRLOd0OhUZGem1FhkZqbq6OosS%2BY68vDyFhYVpypQpVkexhN1uV1BQkGbMmGF1FEvV1NTo/fff12WXXab9%2B/frl7/8pTZs2KC9e/daHe2iULBw0dasWaMZM2aof//%2BVkexlNvtliTdfffdiouLU69evTR37lzt27dPX3/9tcXpOteZM2eUlZWlyZMn69ChQ3rjjTfUs2dP5ebmWh3NJ3z7WME33G631q1bp5KSEhUWFio0NNTqSJ3u%2BPHj%2Bu1vf6sVK1YoKCjI6jiWcrvdiomJ0d13363u3bsrJSVFN910k/bs2WN1tItCwcJFKSsr03vvved1mj9QXX755ZLkdWo/Pj5ebrdbx48ftyqWJcrKynTs2DE99NBD6tmzp%2BLi4pSTk6P//u//ltPptDqepaKjo1vNwOl0KiYmxqJE1nK5XFq0aJH27dunF154Qf369bM6kiXWrl2r6dOnB/y7sCXpiiuuaHX5MD4%2BXg6Hw6JE7WOzOgC6puLiYh0/flw33nijJM%2B/zJOSkvTLX/5SU6dOtTJep%2BrVq5fCw8NVVVWlwYMHS5Kqq6t1ySWXKDY21uJ0nau5uVkul8vrTM2ZM2csTOQ7hgwZooqKCq%2B18vLygHqu/KPVq1frk08%2B0QsvvKCoqCir41imuLhYEREReumllyRJjY2Ncrlcev3117vsu%2Bcu1tVXX60vvvhCDQ0N6tGjh6RvfpbGx8dbnOziULBwURYtWqR58%2Ba1fF9TU6OMjAzt2LGj1etM/J3NZlN6eroef/xxJSYmKjw8XBs3btS0adNkswXWU2zEiBEKCwtTfn6%2BsrKy1NjYqMLCQiUmJgb0X6KSdMcddyg9PV379%2B/XmDFjtHPnTn3%2B%2BedKS0uzOlqnO3z4sIqLi7V79%2B6Af1wcOHDA6/utW7eqpqZGixcvtiiRdVJTUxUREaH//M//1MKFC/XBBx9o7969euqpp6yOdlGC3LwoAAYcO3ZMEyZM0JEjR6yOYokzZ85ozZo12rVrl86ePatJkybpkUceaflXWCCpqKjQo48%2Bqo8//ljdunXT6NGjtWjRIr99h9g/Gjp0qKRv3hUnqaVgf/tOwdLSUq1fv17V1dXq37%2B/li5dqsTERGvCdrBzzWLJkiX6wx/%2B0OofIImJiV36XWM/5HyPi3%2BUn5%2Bv6upqrV27tvMCdqLzzeLPf/6zli9frsrKSsXExGjevHld9qMrKFgAAACG8SJ3AAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADDs/wA39bs%2BUUpLuwAAAABJRU5ErkJggg%3D%3D\"/>\n",
       "        </div>\n",
       "        <div role=\"tabpanel\" class=\"tab-pane col-md-12\" id=\"common8119680243321151841\">\n",
       "            \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">16</td>\n",
       "        <td class=\"number\">71</td>\n",
       "        <td class=\"number\">13.3%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">14</td>\n",
       "        <td class=\"number\">69</td>\n",
       "        <td class=\"number\">12.9%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:97%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">13</td>\n",
       "        <td class=\"number\">61</td>\n",
       "        <td class=\"number\">11.4%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:86%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">8</td>\n",
       "        <td class=\"number\">58</td>\n",
       "        <td class=\"number\">10.8%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:81%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">15</td>\n",
       "        <td class=\"number\">56</td>\n",
       "        <td class=\"number\">10.5%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:79%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">11</td>\n",
       "        <td class=\"number\">55</td>\n",
       "        <td class=\"number\">10.3%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:77%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">3</td>\n",
       "        <td class=\"number\">49</td>\n",
       "        <td class=\"number\">9.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:69%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">9</td>\n",
       "        <td class=\"number\">43</td>\n",
       "        <td class=\"number\">8.0%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:60%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">10</td>\n",
       "        <td class=\"number\">38</td>\n",
       "        <td class=\"number\">7.1%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:53%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">12</td>\n",
       "        <td class=\"number\">35</td>\n",
       "        <td class=\"number\">6.5%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:49%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "        </div>\n",
       "        <div role=\"tabpanel\" class=\"tab-pane col-md-12\"  id=\"extreme8119680243321151841\">\n",
       "            <p class=\"h4\">Minimum 5 values</p>\n",
       "            \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">3</td>\n",
       "        <td class=\"number\">49</td>\n",
       "        <td class=\"number\">9.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:84%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">8</td>\n",
       "        <td class=\"number\">58</td>\n",
       "        <td class=\"number\">10.8%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">9</td>\n",
       "        <td class=\"number\">43</td>\n",
       "        <td class=\"number\">8.0%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:74%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">10</td>\n",
       "        <td class=\"number\">38</td>\n",
       "        <td class=\"number\">7.1%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:65%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">11</td>\n",
       "        <td class=\"number\">55</td>\n",
       "        <td class=\"number\">10.3%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:94%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "            <p class=\"h4\">Maximum 5 values</p>\n",
       "            \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">12</td>\n",
       "        <td class=\"number\">35</td>\n",
       "        <td class=\"number\">6.5%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:49%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">13</td>\n",
       "        <td class=\"number\">61</td>\n",
       "        <td class=\"number\">11.4%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:86%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">14</td>\n",
       "        <td class=\"number\">69</td>\n",
       "        <td class=\"number\">12.9%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:97%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">15</td>\n",
       "        <td class=\"number\">56</td>\n",
       "        <td class=\"number\">10.5%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:79%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">16</td>\n",
       "        <td class=\"number\">71</td>\n",
       "        <td class=\"number\">13.3%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "        </div>\n",
       "    </div>\n",
       "</div>\n",
       "</div><div class=\"row variablerow\">\n",
       "    <div class=\"col-md-3 namecol\">\n",
       "        <p class=\"h4 pp-anchor\" id=\"pp_var_length\">length<br/>\n",
       "            <small>Numeric</small>\n",
       "        </p>\n",
       "    </div><div class=\"col-md-6\">\n",
       "    <div class=\"row\">\n",
       "        <div class=\"col-sm-6\">\n",
       "            <table class=\"stats \">\n",
       "                <tr>\n",
       "                    <th>Distinct count</th>\n",
       "                    <td>530</td>\n",
       "                </tr>\n",
       "                <tr>\n",
       "                    <th>Unique (%)</th>\n",
       "                    <td>99.1%</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Missing (%)</th>\n",
       "                    <td>0.0%</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Missing (n)</th>\n",
       "                    <td>0</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Infinite (%)</th>\n",
       "                    <td>0.0%</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Infinite (n)</th>\n",
       "                    <td>0</td>\n",
       "                </tr>\n",
       "            </table>\n",
       "\n",
       "        </div>\n",
       "        <div class=\"col-sm-6\">\n",
       "            <table class=\"stats \">\n",
       "\n",
       "                <tr>\n",
       "                    <th>Mean</th>\n",
       "                    <td>44474</td>\n",
       "                </tr>\n",
       "                <tr>\n",
       "                    <th>Minimum</th>\n",
       "                    <td>19608</td>\n",
       "                </tr>\n",
       "                <tr>\n",
       "                    <th>Maximum</th>\n",
       "                    <td>143652</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Zeros (%)</th>\n",
       "                    <td>0.0%</td>\n",
       "                </tr>\n",
       "            </table>\n",
       "        </div>\n",
       "    </div>\n",
       "</div>\n",
       "<div class=\"col-md-3 collapse in\" id=\"minihistogram-4137530268564546891\">\n",
       "    <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAMgAAABLCAYAAAA1fMjoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD%2BnaQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAABGVJREFUeJzt3cFLVGsYx/Hf6GXIJCRzRkcCN2FEEERTi2QQN3LHXLSLiFaVbaqVhs7FqRBr6cY2/QGCqyIS7i4oSswLQUIFujBKxmawJpXoONhzdwNi97lNXMep%2B/0s33POnHcxX97D8M5MyMxMAL6parsnAFSy37Z7Av%2BF%2BB9/lnzNX8O/b8FM8KthBQEcBAI4CARwEAjgIBDAQSCAg0AAB4EADgIBHAQCOAgEcBAI4CAQwPFL7Ob9EewAxveouEB%2B5I0LbBUesQAHgQAOAgEcBAI4CARwEAjgIBDAQSCAg0AAB4EADgIBHAQCOAgEcBAI4Ki47e6VrNSt%2BHx/5OfHCgI4CARwEAjgIBDAQSCAg0AAB4EADgIBHAQCOAgEcBAI4CAQwMFmxS3ED2T//AikwhBVZSGQXwDb8LcOgfwPVfJfTFRavCEzs%2B2eBODJZrMaHx/XqVOnFI1Gy3pvPsVCxcvlchodHVUulyv7vQkEcBAI4CAQwEEgqHiRSESXLl1SJBIp%2B735FAtwsIIADgIBHAQCOAgEcBAI4CAQwEEggIPt7iiL5eVlpdNpTU9P68mTJ5Kk/fv369ixY8VzTp8%2Bra6uLs3MzGh4eFjV1dWqrq7WzZs3tXfvXr19%2B1apVErr6%2Bv6%2BvWrBgcHdfDgQX38%2BFH9/f1aWVlRoVDQlStXlEgkFASBUqmUFhYWVCgUdPbsWZ08ebK0iRtQBufOnbOxsTE7fvx4cay1tfWb5yaTSZuenjYzs7t379rFixfNzOz8%2BfN27949MzN79uyZdXd3m5nZtWvX7Pbt22ZmNj8/b4lEwoIgsDt37lgqlTIzs6WlJWtvb7dsNlvSvHnEQlmMjIwokUj863nv3r3T6uqq4vG4JKmrq0uTk5MqFAqamppSMpmUJB09elT5fF6ZTEaPHj3SiRMnJEktLS1qbm7WixcvNozX19fryJEjxdXre/GIhbLYtWuXPn36tGl8YGBAb968USwWU39/v7LZrBoaGorHw%2BGwduzYoQ8fPqimpkbhcLh4LBKJaHFxcdM10Wj0H8ffv39f0rxZQbBt0um0%2Bvr6NDY2ppaWFg0NDZX8GqFQaNOYmZU07iEQbJszZ86ovr5ekpRMJvXq1SvFYjFls9niOZ8/f1YQBNq9e7e%2BfPmiIAiKxxYXFxWLxdTU1LThmkwm446XgkCwLebm5tTT06NCoSBJevr0qQ4cOKBYLKY9e/ZoampKknT//n21t7crHA6rra1NExMTkqTHjx%2BrublZjY2N6ujo0IMHDyRJs7OzWlpa0qFDhzaM53I5PX/%2BXG1tbSXNk%2B3u2HL5fF6XL19WEAR6%2BfKlDh8%2BrNbWVtXV1enhw4eqra3Vzp07NTQ0pMbGRr1%2B/Vo3btxQKBRSTU2Nbt26pWg0qkwmo4GBAa2tramqqkrXr1/Xvn37tLKyoqtXryqfz8vM1Nvbq3g8rrW1NaXTac3Pz2t9fV0XLlxQZ2dnSXMnEMDBIxbgIBDAQSCA429Unz2M9PTtkgAAAABJRU5ErkJggg%3D%3D\">\n",
       "\n",
       "</div>\n",
       "<div class=\"col-md-12 text-right\">\n",
       "    <a role=\"button\" data-toggle=\"collapse\" data-target=\"#descriptives-4137530268564546891,#minihistogram-4137530268564546891\"\n",
       "       aria-expanded=\"false\" aria-controls=\"collapseExample\">\n",
       "        Toggle details\n",
       "    </a>\n",
       "</div>\n",
       "<div class=\"row collapse col-md-12\" id=\"descriptives-4137530268564546891\">\n",
       "    <ul class=\"nav nav-tabs\" role=\"tablist\">\n",
       "        <li role=\"presentation\" class=\"active\"><a href=\"#quantiles-4137530268564546891\"\n",
       "                                                  aria-controls=\"quantiles-4137530268564546891\" role=\"tab\"\n",
       "                                                  data-toggle=\"tab\">Statistics</a></li>\n",
       "        <li role=\"presentation\"><a href=\"#histogram-4137530268564546891\" aria-controls=\"histogram-4137530268564546891\"\n",
       "                                   role=\"tab\" data-toggle=\"tab\">Histogram</a></li>\n",
       "        <li role=\"presentation\"><a href=\"#common-4137530268564546891\" aria-controls=\"common-4137530268564546891\"\n",
       "                                   role=\"tab\" data-toggle=\"tab\">Common Values</a></li>\n",
       "        <li role=\"presentation\"><a href=\"#extreme-4137530268564546891\" aria-controls=\"extreme-4137530268564546891\"\n",
       "                                   role=\"tab\" data-toggle=\"tab\">Extreme Values</a></li>\n",
       "\n",
       "    </ul>\n",
       "\n",
       "    <div class=\"tab-content\">\n",
       "        <div role=\"tabpanel\" class=\"tab-pane active row\" id=\"quantiles-4137530268564546891\">\n",
       "            <div class=\"col-md-4 col-md-offset-1\">\n",
       "                <p class=\"h4\">Quantile statistics</p>\n",
       "                <table class=\"stats indent\">\n",
       "                    <tr>\n",
       "                        <th>Minimum</th>\n",
       "                        <td>19608</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>5-th percentile</th>\n",
       "                        <td>24963</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Q1</th>\n",
       "                        <td>32426</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Median</th>\n",
       "                        <td>41443</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Q3</th>\n",
       "                        <td>52934</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>95-th percentile</th>\n",
       "                        <td>75740</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Maximum</th>\n",
       "                        <td>143652</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Range</th>\n",
       "                        <td>124044</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Interquartile range</th>\n",
       "                        <td>20509</td>\n",
       "                    </tr>\n",
       "                </table>\n",
       "            </div>\n",
       "            <div class=\"col-md-4 col-md-offset-2\">\n",
       "                <p class=\"h4\">Descriptive statistics</p>\n",
       "                <table class=\"stats indent\">\n",
       "                    <tr>\n",
       "                        <th>Standard deviation</th>\n",
       "                        <td>16451</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Coef of variation</th>\n",
       "                        <td>0.3699</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Kurtosis</th>\n",
       "                        <td>4.105</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Mean</th>\n",
       "                        <td>44474</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>MAD</th>\n",
       "                        <td>12459</td>\n",
       "                    </tr>\n",
       "                    <tr class=\"\">\n",
       "                        <th>Skewness</th>\n",
       "                        <td>1.5343</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Sum</th>\n",
       "                        <td>23793475</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Variance</th>\n",
       "                        <td>270640000</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Memory size</th>\n",
       "                        <td>4.3 KiB</td>\n",
       "                    </tr>\n",
       "                </table>\n",
       "            </div>\n",
       "        </div>\n",
       "        <div role=\"tabpanel\" class=\"tab-pane col-md-8 col-md-offset-2\" id=\"histogram-4137530268564546891\">\n",
       "            <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAlgAAAGQCAYAAAByNR6YAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD%2BnaQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3XtYlOW%2B//EPMEoqoGKBRSyzPIumKVJqnjOzVXkMXbvMsjxskjRtoVZr67YEt7jIpCxt19LsQLnaS7NMU8us1V5lWeIhy0OXhzAmZeSQKIf794c/ZzsyCMgzzsC8X9flRdz3Mzff%2BXo7feZ5ZoYAY4wRAAAALBPo7QIAAABqGwIWAACAxQhYAAAAFiNgAQAAWIyABQAAYDECFgAAgMUIWAAAABYjYAEAAFiMgAUAAGAxAhYAAIDFCFgAAAAWI2ABAABYjIAFAABgMQIWAACAxQhYAAAAFiNgAQAAWIyABQAAYDECFgAAgMUIWAAAABYjYAEAAFiMgAUAAGAxAhYAAIDFCFgAAAAWI2ABAABYjIAFAABgMQIWAACAxQhYAAAAFiNgAQAAWIyABQAAYDECFgAAgMUIWAAAABYjYAEAAFiMgAUAAGAxAhYAAIDFCFgAAAAWI2ABAABYjIAFAABgMQIWAACAxQhYAAAAFrN5uwB/YbfneWTdwMAAhYc30IkTBSotNR75GTUVvXGPvrhHX8pHb9yjL%2B75Wl%2BuuirUKz%2BXM1g1XGBggAICAhQYGODtUnwOvXGPvrhHX8pHb9yjL%2B7Rl7MIWAAAABYjYAEAAFiMgAUAAGAxAhYAAIDFCFgAAAAWI2ABAABYjIAFAABgMQIWAACAxQhYAAAAFiNgAQAAWIyABQAAYDECFgAAgMUIWAAAABazebsA%2BI87nvvC2yVUybopPbxdAgCghuIMFgAAgMUIWAAAABYjYAEAAFiMgAUAAGAxAhYAAIDFCFgAAAAWq7Uf07B161YlJSUpLi5OaWlpzvGnnnpKq1evdjm2pKRE99xzj5KTkzVjxgytWbNGQUFBzvng4GBt27btstUOAABqtloZsJYtW6ZVq1apWbNmZeaeeeYZPfPMM87vi4uLNWTIEA0aNMg5NmnSJE2ePPmy1AoAAGqfWnmJMDg4uNyAdaHly5frmmuuUe/evS9DZQAAwB/UyoA1ZswYhYaGVnhcbm6uXnrpJT3xxBMu4//7v/%2BrIUOGqHPnzhoxYoR27tzpqVIBAEAtVCsvEVbWypUrFRsbq5YtWzrHoqOjFRgYqMcee0wNGjRQenq6HnroIa1fv16NGzeu1LrZ2dmy2%2B0uYzZbfUVERFhavyQFBQW6fIV1bLba2VP2jHv0pXz0xj364h59OctvA1ZJSYneeOMNLVy40GU8ISHB5fsnnnhCa9eu1caNGzVy5MhKrZ2RkaH09PQy6yYmJlav6IsIC6vnsbX9VePGDbxdgkexZ9yjL%2BWjN%2B7RF/f8vS9%2BG7C%2B/vprnTlzRl27dr3ocUFBQbr66quVnZ1d6bXj4%2BPVr18/lzGbrb5ycgouqdaLCQoKVFhYPeXmnlJJSanl6/szT/x9%2BQL2jHv0pXz0xj364p6v9cVbT5b9NmBt2rRJN998s2y2/2uBMUYpKSkaOnSo2rRpI0k6c%2BaMDh06pOjo6EqvHRERUeZyoN2ep%2BJiz220kpJSj67vj2p7P9kz7tGX8tEb9%2BiLe/7eF7%2B9QLpnzx5de%2B21LmMBAQE6cuSI5syZo19//VUFBQVKTU1VnTp1NGDAAC9VCgAAappaeQarQ4cOks5%2BxpUkbdy4UZKUmZnpPMZut%2BvKK68sc9tnn31W8%2BfP17Bhw5Sfn6%2BOHTtq%2BfLlql%2B//mWoHAAA1Aa1MmCdH6TKs379erfjjRo1UnJystUlAQAAP%2BK3lwgBAAA8hYAFAABgMQIWAACAxQhYAAAAFiNgAQAAWIyABQAAYDECFgAAgMUIWAAAABYjYAEAAFiMgAUAAGAxAhYAAIDFCFgAAAAWI2ABAABYjIAFAABgMQIWAACAxQhYAAAAFiNgAQAAWIyABQAAYDECFgAAgMUIWAAAABYjYAEAAFiMgAUAAGAxAhYAAIDFCFgAAAAWI2ABAABYjIAFAABgMQIWAACAxQhYAAAAFiNgAQAAWIyABQAAYDECFgAAgMVqbcDaunWrunfvrqlTp7qMv/fee2rTpo06dOjg8mfHjh2SpNLSUqWlpal///6KjY3VuHHjdPjwYW/cBQAAUEPZvF2AJyxbtkyrVq1Ss2bN3M7Hxsbq9ddfdzv3xhtv6P3339eyZcsUGRmptLQ0JSQkaPXq1QoICPBk2QAAoJaolWewgoODLxqwLiYjI0Njx47VDTfcoJCQEE2dOlX79%2B/X999/74FKAQBAbVQrA9aYMWMUGhpa7nxWVpYefPBBxcbGqn///lq9erUkqbCwUPv27VO7du2cx4aEhKhZs2bKzMz0eN0AAKB2qJWXCC8mPDxc1113nR5//HG1aNFCH3/8sf785z8rIiJC119/vYwxatiwocttGjZsqJycnEr/jOzsbNntdpcxm62%2BIiIiLLkP5wsKCnT5CuvYbLWzp%2BwZ9%2BhL%2BeiNe/TFPfpylt8FrD59%2BqhPnz7O7%2B%2B88059/PHHeu%2B99zR9%2BnRJkjGmWj8jIyND6enpLmMJCQlKTEys1roXExZWz2Nr%2B6vGjRt4uwSPYs%2B4R1/KR2/coy/u%2BXtf/C5guRMVFaWdO3eqUaNGCgwMlMPhcJl3OBxq0qRJpdeLj49Xv379XMZstvrKySmwpN7zBQUFKiysnnJzT6mkpNTy9f2ZJ/6%2BfAF7xj36Uj564x59cc/X%2BuKtJ8t%2BF7DeeustNWzYUIMHD3aO7d%2B/X9HR0QoODlbLli21a9cudevWTZKUm5urQ4cOqWPHjpX%2BGREREWUuB9rteSou9txGKykp9ej6/qi295M94x59KR%2B9cY%2B%2BuOfvffG7C6RnzpzR3LlzlZmZqaKiIq1du1afffaZRo0aJUkaPXq0VqxYof379ys/P1%2Bpqalq27atOnTo4OXKAQBATVErz2CdC0PFxcWSpI0bN0qSMjMzNWbMGBUUFOixxx6T3W7XtddeqxdeeEExMTGSpFGjRslut%2Bv%2B%2B%2B9XQUGB4uLiyryeCgAA4GICTHVf0Y1KsdvzPLKuzRaoxo0bKCenwOdPxd7x3BfeLqFK1k3p4e0SPKIm7ZnLib6Uj964R1/c87W%2BXHVV%2BR/b5El%2Bd4kQAADA0whYAAAAFiNgAQAAWIyABQAAYDECFgAAgMUIWAAAABYjYAEAAFiMgAUAAGAxAhYAAIDFCFgAAAAWI2ABAABYjIAFAABgMQIWAACAxQhYAAAAFiNgAQAAWIyABQAAYDECFgAAgMUIWAAAABYjYAEAAFiMgAUAAGAxAhYAAIDFCFgAAAAWI2ABAABYjIAFAABgMQIWAACAxQhYAAAAFiNgAQAAWIyABQAAYDECFgAAgMUIWAAAABYjYAEAAFjM5u0CPGXr1q1KSkpSXFyc0tLSXOY2bNig9PR0HT58WBERERo3bpzuvfdeSdLixYv14osvymZzbc0nn3yiK6%2B88rLVD%2B%2B747kvvF1Clayb0sPbJQAA/r9aGbCWLVumVatWqVmzZmXmduzYoenTp%2Buvf/2r%2BvTpoy%2B%2B%2BEIJCQm6/vrr1bVrV0nSPffco5SUlMtdNgAAqCVq5SXC4ODgcgOWw%2BHQhAkTNGDAANlsNvXu3VutWrXStm3bvFApAACojWrlGawxY8aUO9erVy/16tXL%2BX1xcbHsdrsiIyOdY3v37tWoUaP0448/6uqrr9bMmTPVs2dPj9Z8qW5L3ertEgAAwAVqZcCqitTUVNWvX1%2BDBw%2BWJDVt2lTR0dGaNm2aIiIilJGRoYkTJ2rNmjW6/vrrK7Vmdna27Ha7y5jNVl8RERGW1w%2BcY7NV7oR0UFCgy1ecRV/KR2/coy/u0Zez/DZgGWOUmpqqtWvXasWKFQoODpYkjRw5UiNHjnQeN3bsWH3wwQdas2aNpkyZUqm1MzIylJ6e7jKWkJCgxMRE6%2B4AcIHGjRtU6fiwsHoeqqRmoy/lozfu0Rf3/L0vfhmwSktLNXPmTO3YsUNvvfWWoqOjL3p8VFSUsrOzK71%2BfHy8%2BvXr5zJms9VXTk7BJdULVEZl91dQUKDCwuopN/eUSkpKPVxVzUFfykdv3KMv7vlaX6r65NMqfhmw5s2bp59%2B%2BklvvfWWGjVq5DL34osvqnPnzrrlllucY/v373deQqyMiIiIMpcD7fY8FRd7f6Oh9qrq/iopKWVPukFfykdv3KMv7vl7X/zuAuk333yjNWvWaOnSpWXClXT2XYZz5szRgQMHdPr0ab366qs6dOiQhg4d6oVqAQBATVQrz2B16NBB0tl3CErSxo0bJUmZmZn6%2B9//rry8PPXt29flNrGxsXr11Vc1bdo0SWdfe%2BVwONSiRQv97W9/U9OmTS/jPQAAADVZgDHGeLsIf2C353lk3Zr2aePwnMp%2BkrvNFqjGjRsoJ6fAr0/fX4i%2BlI/euEdf3PO1vlx1VahXfq7fXSIEAADwNAIWAACAxQhYAAAAFiNgAQAAWIyABQAAYDECFgAAgMUIWAAAABYjYAEAAFiMgAUAAGAxAhYAAIDFfC5g9evXT%2Bnp6crKyvJ2KQAAAJfE5wLW8OHD9eGHH2rAgAF6%2BOGHtWHDBucvbQYAAKgJfC5gJSQk6MMPP9Q777yjli1bat68eerdu7cWLFiggwcPers8AACACvlcwDqnffv2SkpK0ieffKJZs2bpnXfe0eDBgzVu3Djt2LHD2%2BUBAACUy2cDVlFRkT788EM98sgjSkpKUmRkpGbOnKm2bdtq7Nixev/9971dIgAAgFs2bxdwof3792vVqlX6xz/%2BoYKCAt1%2B%2B%2B1avny5unTp4jwmNjZWs2fP1l133eXFSgEAANzzuYB15513qnnz5powYYKGDBmiRo0alTmmd%2B/eOnHihBeqAwAAqJjPBawVK1aoW7duFR73/fffX4ZqAAAAqs7nXoPVunVrTZw4URs3bnSO/e1vf9Mjjzwih8PhxcoAAAAqx%2BcCVnJysvLy8tSiRQvnWJ8%2BfVRaWqqUlBQvVgYAAFA5PneJ8PPPP9f777%2Bvxo0bO8euu%2B46paam6o9//KMXKwMAAKgcnzuDVVhYqODg4DLjgYGBOnXqlBcqAgAAqBqfC1ixsbFKSUnRyZMnnWO//vqr5syZ4/JRDQAAAL7K5y4Rzpo1Sw899JBuueUWhYSEqLS0VAUFBYqOjtbrr7/u7fIAAAAq5HMBKzo6Wh988IE%2B%2B%2BwzHTp0SIGBgWrevLl69uypoKAgb5cHAABQIZ8LWJJUt25dDRgwwNtlAAAAXBKfC1iHDx/WwoUL9dNPP6mwsLDM/KZNm7xQFQAAQOX5XMCaNWuWsrOz1bNnT9WvX9/b5QAAAFSZzwWsnTt3atOmTQoPD/d2KQAAAJfE5z6moUmTJpy5AgAANZrPBawJEyYoPT1dxhhvlwIAAHBJfO4S4WeffaZvv/1W7733nq699loFBrpmwLfffrtS62zdulVJSUmKi4tTWlqay9yHH36oJUuW6MiRI2revLkef/xx9ezZU5JUWlqqRYsWae3atcrNzVXHjh01e/ZsRUdHW3MHAQBAredzASskJES9evWq1hrLli3TqlWr1KxZszJze/bsUVJSktLT03XzzTdr/fr1evTRR/XRRx%2BpadOmeuONN/T%2B%2B%2B9r2bJlioyMVFpamhISErR69WoFBARUqy4AAOAffC5gJScnV3uN4OBgrVq1Ss8%2B%2B6xOnz7tMvfuu%2B%2Bqd%2B/e6t27tyTp7rvv1sqVK7VmzRqNHz9eGRkZGjt2rG644QZJ0tSpUxUXF6fvv/9enTp1qnZtAACg9vO512BJ0oEDB7R48WLNnDnTObZ9%2B/ZK337MmDEKDQ11O7dr1y61a9fOZaxdu3bKzMxUYWGh9u3b5zIfEhKiZs2aKTMzs4r3AgAA%2BCufO4P15Zdf6pFHHlHz5s31888/Kzk5WYcPH9aYMWP03HPPqX///tVa3%2BFwqGHDhi5jDRs21L59%2B3Ty5EkZY9zO5%2BTkVPpnZGdny263u4zZbPUVERFx6YUDFbDZKvd8KSgo0OUrzqIv5aM37tEX9%2BjLWT4XsNLS0vTEE0/ogQceUMeOHSWd/f2EKSkpeuGFF6odsCRV%2BA7F6r6DMSMjQ%2Bnp6S5jCQkJSkxMrNa6wMU0btygSseHhdXzUCU1G30pH71xj7645%2B998bmA9eOPP2rlypWS5PKi8kGDBmnWrFnVXr9x48ZyOBwuYw6HQ%2BHh4WrUqJECAwPdzjdp0qTSPyM%2BPl79%2BvVzGbPZ6isnp%2BDSCwcqUNn9FRQUqLCwesrNPaWSklIPV1Vz0Jfy0Rv36It7vtaXqj75tIrPBazQ0FAVFhaqbt26LuPZ2dllxi5FTEyMdu7c6TKWmZmpO%2B%2B8U8HBwWrZsqV27dqlbt26SZJyc3N16NAh59m0yoiIiChzOdBuz1Nxsfc3Gmqvqu6vkpJS9qQb9KV89MY9%2BuKev/fF5y6Q3nTTTZo3b57y8/OdYwcPHlRSUpJuueWWaq9/77336p///Kc%2B/fRTnT59WqtWrdLPP/%2Bsu%2B%2B%2BW5I0evRorVixQvv371d%2Bfr5SU1PVtm1bdejQodo/GwAA%2BAefO4M1c%2BZMPfDAA4qLi1NJSYluuukmnTp1Si1btlRKSkql1jgXhoqLiyVJGzdulHT2TFWrVq2Umpqq5ORkHT16VC1atNDLL7%2Bsq666SpI0atQo2e123X///SooKFBcXFyZ11MBAABcTIDxwd9JU1RUpC1btujgwYO64oor1Lx5c/Xo0aNGf9Cn3Z7nkXXveO4Lj6yLmmfdlB6VOs5mC1Tjxg2Uk1Pg16fvL0Rfykdv3KMv7vlaX666yv3HNnmaz53BkqQ6depowIAB3i4DAADgkvhcwOrXr99Fz1Rt2rTpMlYDAABQdT4XsAYPHuwSsEpKSnTw4EFlZmbqgQce8GJlAAAAleNzAWv69Olux9evX69//etfl7kaAACAqvO5j2koz4ABA/TBBx94uwwAAIAK1ZiAtXv37mr/ChsAAIDLwecuEY4aNarM2KlTp7R//34NHDjQCxUBAABUjc8FrOuuu67MuwiDg4M1YsQIjRw50ktVAQAAVJ7PBazKflo7AACAr/K5gPWPf/yj0scOGTLEg5UAAABcGp8LWE8%2B%2BaRKS0vLvKA9ICDAZSwgIICABQAAfJLPBaxXXnlFr776qiZOnKjWrVvLGKO9e/dq2bJluu%2B%2B%2BxQXF%2BftEgEAAC7K5wJWSkqKli5dqsjISOdY165dFR0drXHjxmnt2rVerA4AAKBiPvc5WD///LMaNmxYZjwsLExHjx71QkUAAABV43MBKyoqSikpKcrJyXGO5ebmauHChfrDH/7gxcoAAAAqx%2BcuEc6aNUvTpk1TRkaGGjRooMDAQOXn5%2BuKK67QCy%2B84O3yAAAAKuRzAatnz5769NNPtWXLFh07dkzGGEVGRurWW29VaGiot8sDAACokM8FLEmqV6%2Be%2Bvfvr2PHjik6Otrb5QAAAFSJz70Gq7CwUElJSercubPuuOMOSWdfg/Xwww8rNzfXy9UBAABUzOcC1oIFC7Rnzx6lpqYqMPD/yispKVFqaqoXKwMAAKgcnwtY69ev1/PPP69BgwY5f%2BlzWFiYkpOTtWHDBi9XBwAAUDGfC1gFBQW67rrryoyHh4fr999/v/wFAQAAVJHPBaw//OEP%2Bte//iVJLr978KOPPtI111zjrbIAAAAqzefeRfinP/1JkydP1vDhw1VaWqrXXntNO3fu1Pr16/Xkk096uzwAAIAK%2BVzAio%2BPl81m08qVKxUUFKSXXnpJzZs3V2pqqgYNGuTt8gAAACrkcwHrxIkTGj58uIYPH%2B7tUgAAAC6Jz70Gq3///i6vvQIAAKhpfC5gxcXFad26dd4uAwAA4JL53CXCq6%2B%2BWs8%2B%2B6yWLl2qP/zhD6pTp47L/MKFC71UGQAAQOX4XMDat2%2Bfrr/%2BeklSTk6Ol6sBAACoOp8JWFOnTlVaWppef/1159gLL7yghIQEL1YFAABQdT4TsDZv3lxmbOnSpR4JWF9//bUeeughlzFjjIqKirRixQqNGTNGdevWdZn/r//6L%2BcvnwYAALgYnwlY7t456Kl3E8bGxiozM9Nl7KWXXtIPP/wgSYqKinIb%2BAAAACrDZwLWuV/sXNGYJ/zyyy967bXX9D//8z86fPjwZfmZAACg9vKZgOVNixYt0vDhw3XNNdfo8OHDKigoUEJCgrZt26a6devqoYce0tixYysd%2BLKzs2W3213GbLb6ioiI8ET5gCTJZqvcp64EBQW6fMVZ9KV89MY9%2BuIefTnL7wPWkSNHtGHDBm3YsEGSFBISolatWumBBx5QWlqavvrqKz322GMKDQ3ViBEjKrVmRkaG0tPTXcYSEhKUmJhoef3AOY0bN6jS8WFh9TxUSc1GX8pHb9yjL%2B75e18CjI98bHq7du3KvIh83bp1Zcas/hys%2BfPn68SJE5o/f365xyxYsEDbt2/Xm2%2B%2BWak1L%2BcZrNtSt1q%2BJmqmj6ffWqnjgoICFRZWT7m5p1RSUurhqmoO%2BlI%2BeuMefXHP1/pS1SefVvGZM1hdunRRdnZ2hWNWW79%2BvZKSki56TFRUlNavX1/pNSMiIsqEKbs9T8XF3t9oqL2qur9KSkrZk27Ql/LRG/foi3v%2B3hefCVjnf/7V5bJnzx4dPXpUPXr0cI6tW7dOOTk5%2BtOf/uQcO3DggKKjoy97fQAAoGby61eg7d69W40aNVJISIhzrE6dOpo/f74%2B//xzFRUV6YsvvtDf//53jR492ouVAgCAmsRnzmB5w2%2B//aarrrrKZWzAgAGaNWuW5s6dq6ysLF155ZWaNWuWBg4c6KUqAQBATePXAWvChAmaMGFCmfH4%2BHjFx8d7oSIAAFAb%2BPUlQgAAAE8gYAEAAFiMgAUAAGAxAhYAAIDFCFgAAAAWI2ABAABYjIAFAABgMQIWAACAxQhYAAAAFiNgAQAAWMyvf1UOUJvc8dwX3i6h0tZN6eHtEgDAoziDBQAAYDECFgAAgMUIWAAAABYjYAEAAFiMgAUAAGAxAhYAAIDFCFgAAAAWI2ABAABYjIAFAABgMQIWAACAxQhYAAAAFiNgAQAAWIyABQAAYDECFgAAgMUIWAAAABYjYAEAAFiMgAUAAGAxAhYAAIDFCFgAAAAWI2ABAABYzObtAryhdevWqlOnjgICApxj9957r55%2B%2Bml9%2BeWXWrhwoQ4cOKCrr75aEyZM0N133%2B3FagEAQE3jlwFLkj766CNde%2B21LmPZ2dn693//dz355JO666679M0332jSpElq3ry5OnTo4KVKAQBATcMlwvO8//77uu666zRixAgFBwere/fu6tevn959911vlwYAAGoQvz2DtXDhQm3fvl35%2Bfm64447NGPGDO3atUvt2rVzOa5du3Zat25dldbOzs6W3W53GbPZ6isiIqLadQO1gc3me8/tgoICXb7i/9Ab9%2BiLe/TlLL8MWJ06dVL37t01f/58HT58WFOmTNGcOXPkcDgUGRnpcmyjRo2Uk5NTpfUzMjKUnp7uMpaQkKDExMRq1w7UBo0bN/B2CeUKC6vn7RJ8Fr1xj7645%2B998cuAlZGR4fzvG264QdOnT9ekSZPUpUsXS9aPj49Xv379XMZstvrKySmwZH2gpvPFfwtBQYEKC6un3NxTKikp9XY5PoXeuEdf3PO1vnjrCZ1fBqwLXXvttSopKVFgYKAcDofLXE5OjsLDw6u0XkRERJnLgXZ7noqLvb/RAF/gy/8WSkpKfbo%2Bb6I37tEX9/y9L353gXT37t1KSUlxGdu/f7/q1q2r3r17a%2BfOnS5zO3fu1I033ng5SwQAADWc3wWsJk2aKCMjQ0uXLtWZM2d08OBBLVq0SPHx8brnnnt09OhRvfvuuzp9%2BrS2bNmiLVu26N577/V22QAAoAbxu4AVGRmppUuXavPmzYqLi9OoUaN066236oknnlCTJk308ssva%2BXKlerSpYvmzZunBQsWqE2bNt4uGwAA1CB%2B%2BRqs2NhYvf322%2BXOrV69%2BjJXBAAAahO/O4MFAADgaQQsAAAAixGwAAAALEbAAgAAsBgBCwAAwGIELAAAAIsRsAAAACxGwAIAALAYAQsAAMBiBCwAAACLEbAAAAAsRsACAACwGAELAADAYgQsAAAAixGwAAAALEbAAgAAsBgBCwAAwGIELAAAAIsRsAAAACxGwAIAALAYAQsAAMBiBCwAAACLEbAAAAAsRsACAACwGAELAADAYgQsAAAAixGwAAAALEbAAgAAsBgBCwAAwGIELAAAAIsRsAAAACxm83YB3nL06FHNmzdP27ZtU1BQkHr16qVZs2YpNzdX/fv3V926dV2OnzJlisaNG%2BelagEAQE3itwFr4sSJiomJ0ebNm5WXl6eEhATNnz9fkyZNkiRlZmZ6uUIAAFBT%2BeUlwtzcXMXExGjatGlq0KCBmjZtqqFDh2rbtm3eLg0AANQCfnkGKywsTMnJyS5jWVlZioiIcH7/5z//Wf/85z9VXFyskSNHKjExUXXq1KnU%2BtnZ2bLb7S5jNlt9l/UBf2az%2Bd5zu6CgQJev%2BD/0xj364h59OcsvA9aFMjMztXLlSi1ZskR169ZV586dddttt%2BnZZ5/Vnj17NHnyZNlsNj322GOVWi8jI0Pp6ekuYwkJCUpMTPRE%2BUCN07hxA2%2BXUK6wsHreLsFn0Rv36It7/t6XAGOM8XYR3vTNN99o0qRJevTRRzVmzBi3x7z55pt6%2BeWXtWXLlkqteTnPYN2WutXyNQFP%2B3j6rd4uoYygoECFhdVTbu4plZSUerscn0Jv3KMv7vlaX7z1hM6vz2Bt3rxZTzzxhJ5%2B%2BmkNGTKk3OOioqL022%2B/yRijgICACteNiIgoE6bs9jwVF3t/owG%2BwJf/LZSUlPp0fd5Eb9yjL%2B75e1/89gLpt99%2Bq6SkJC1atMglXH355ZdasmSJy7EHDhxQVFRUpcIVAACAXwas4uJiPfXUU5o%2Bfbp69uzpMhcaGqoXXnhBq1evVlFRkTIzM/Xf//3fGj16tJcoCNlOAAAPc0lEQVSqBQAANY1fXiL87rvvtH//fj3zzDN65plnXOY%2B%2BugjpaWlKT09XX/5y18UGhqq%2B%2B%2B/Xw888ICXqgUAADWNXwasrl27au/eveXOR0VF6bbbbruMFQEAgNrELy8RAgAAeBIBCwAAwGIELAAAAIsRsAAAACxGwAIAALAYAQsAAMBiBCwAAACLEbAAAAAsRsACAACwGAELAADAYgQsAAAAixGwAAAALEbAAgAAsJjN2wUA8D93PPeFt0uoknVTeni7BAA1DGewAAAALMYZLACoAGfcAFQVZ7AAAAAsRsACAACwGAELAADAYgQsAAAAixGwAAAALEbAAgAAsBgBCwAAwGIELAAAAIsRsAAAACxGwAIAALAYAQsAAMBi/C5CAKhlatLvTuT3JqK24gwWAACAxQhYAAAAFiNgAQAAWIyA5cbRo0c1fvx4xcXFqW/fvlqwYIFKS0u9XRYAAKgheJG7G5MnT1b79u21ceNGHT9%2BXBMmTNCVV16pBx980NulAUCtUpNekC/xonxUHgHrApmZmfrhhx/02muvKTQ0VKGhoRo7dqyWL19OwAIAP1fTAmFNUtvCKwHrArt27VJUVJQaNmzoHGvfvr0OHjyo/Px8hYSEVLhGdna27Ha7y5jNVl8RERGW1wsAQG1gs9WuVy0RsC7gcDgUFhbmMnYubOXk5FQqYGVkZCg9Pd1l7NFHH9XkyZOtK/T/%2B/Cxm5SRkaH4%2BHgC3AWys7PpjRv0xT36Uj564x59cY%2B%2BnFW74qJFjDHVun18fLzee%2B89lz/x8fEWVefKbrcrPT29zBkz0Jvy0Bf36Ev56I179MU9%2BnIWZ7AuEB4eLofD4TLmcDgUEBCg8PDwSq0RERHh16kdAAB/xxmsC8TExCgrK0snTpxwjmVmZqpFixZq0KCBFysDAAA1BQHrAu3atVOHDh20cOFC5efna//%2B/Xrttdc0evRob5cGAABqiKDZs2fP9nYRvubWW2/V2rVrNXfuXH3wwQcaMWKExo0bp4CAAG%2BX5laDBg3UrVs3zrC5QW/coy/u0Zfy0Rv36It79EUKMNV9RTcAAABccIkQAADAYgQsAAAAixGwAAAALEbAAgAAsBgBCwAAwGIELAAAAIsRsAAAACxGwAIAALAYAQsAAMBiBCwvO3r0qBISEhQXF6fu3btrxowZys3NlSTt2bNH9913n7p06aKBAwfq1Vdfdbnthx9%2BqLvuukudO3fWsGHD9PnnnzvnSktLlZaWpv79%2Bys2Nlbjxo3T4cOHnfMOh0NTpkxR9%2B7d1bNnTz355JMqLCy8PHe6iubNm6fWrVs7v//yyy81YsQI3XTTTbrzzju1Zs0al%2BNXrFih22%2B/XTfddJNGjx6tnTt3OudOnz6tv/zlL%2BrVq5fi4uKUmJionJwc5/zRo0c1fvx4xcXFqW/fvlqwYIFKS0s9fyerYMmSJerZs6c6deqksWPH6siRI5Loy%2B7duzVmzBh17dpVPXr00PTp052/tN2ferN161Z1795dU6dOLTPnyceM6jxeXS4X682GDRt09913q3Pnzrr99tv1zjvvuMx7co9UtD897WJ9OaegoEB9%2BvTRjBkznGP%2BsGeqxcCr/vjHP5oZM2aY/Px8k5WVZYYNG2ZmzZplTp06ZW699VazePFiU1BQYHbu3Gm6detm1q9fb4wxZvfu3SYmJsZ8%2BumnprCw0KxevdrceOONJisryxhjzIoVK0zfvn3Nvn37TF5envnP//xPc9ddd5nS0lJjjDGPPvqoGT9%2BvDl%2B/Lg5duyYiY%2BPN3PnzvVaH8qze/du061bN9OqVStjjDG//vqr6dSpk3n33XdNYWGh%2BeKLL0zHjh3Njh07jDHGbNq0yXTt2tV899135tSpU%2Bbll182PXr0MAUFBcYYY5KTk82wYcPML7/8YnJycsyjjz5qJkyY4Px5Q4cONU899ZTJzc01Bw8eNAMHDjSvvvrq5b/j5Vi5cqUZNGiQ2b9/v8nLyzNz5841c%2BfO9fu%2BFBUVmR49epiFCxea06dPmxMnTpgHH3zQTJ482a96s3TpUjNw4EAzatQoM2XKFJc5Tz5mVPfxytu9%2Bf77702HDh3Mxx9/bIqKisynn35q2rdvb77%2B%2BmtjjGf3SEX705t9OV9ycrLp0qWLSUpKco7V9j1TXQQsLzp58qSZMWOGsdvtzrHXX3/dDBw40Kxbt87cfPPNpri42Dm3YMEC89BDDxljjJkzZ45JSEhwWW/kyJHm5ZdfNsYYc%2Bedd5rly5c75/Ly8ky7du3M9u3bjd1uN23atDF79uxxzm/ZssV06tTJnDlzxiP39VKUlJSYkSNHmhdffNEZsF555RUzZMgQl%2BOmTJlinn76aWOMMePHjzfz5s1zWaNHjx5m7dq1pqioyHTp0sVs3LjROb9v3z7TunVrc%2BzYMbNjxw7Ttm1b43A4nPNvvvmmuf322z15N6ukX79%2Bzgeg8/l7X3755RfTqlUrs2/fPufYm2%2B%2BaQYMGOBXvVm%2BfLnJzc01SUlJZf5n6cnHjOo%2BXl0OF%2BvNli1bTHp6usvY0KFDzZIlS4wxnt0jFe1PT7tYX87Zs2eP6dGjh3nmmWdcAlZt3zPVxSVCLwoLC1NycrKuvPJK51hWVpYiIiK0a9cutW7dWkFBQc65du3aOU9L79q1S%2B3atXNZr127dsrMzFRhYaH27dvnMh8SEqJmzZopMzNTe/bsUVBQkMtlt/bt2%2Bv333/XgQMHPHV3q%2Bztt99WcHCw7rrrLudYefe7vL4EBgaqbdu2yszM1KFDh5SXl6f27ds752%2B44QZdccUV2rVrl3bt2qWoqCg1bNjQOd%2B%2BfXsdPHhQ%2Bfn5nrqblfbrr7/qyJEjOnnypAYPHuy8FHHixAm/7oskRUZGqm3btsrIyFBBQYGOHz%2BuDRs2qE%2BfPn7VmzFjxig0NNTtnCcfM6rzeHW5XKw3vXr1UkJCgvP74uJi2e12RUZGSvLsHqlof3raxfoiScYYzZ49W1OnTlVYWJhz3B/2THURsHxIZmamVq5cqUmTJsnhcLhsZklq1KiRHA6HSktL5XA4XP7BSlLDhg2Vk5OjkydPyhhT7rzD4VBISIgCAgJc5iS5vG7Am3777TctXrxY//Ef/%2BEyXl5fztV9sb44HA5JKnP7sLAw5/yFc77Ul2PHjkmSPvroI7322mtavXq1jh07pqeeesqv%2ByKd/R/e4sWLtWnTJt10003q3r27iouLNW3aNL/vzTmefMyozuOVL0pNTVX9%2BvU1ePBgSZ7dIxXtT2/LyMhQQECAhg0b5jLOnqkYActHfPPNNxo3bpymTZum7t27l3vc%2BZvVGHPRNS82X9FtvS05OVnDhg1TixYtqnzb2tqXc7U9/PDDioyMVNOmTTV58mRt3ry5Sre/lHlf7osknTlzRhMnTtSgQYO0bds2ffbZZwoNDdX06dMrdfva3JvzXe77WZXHK19gjNGCBQu0du1aLVmyRMHBwS5zFd32UuZ82fHjx7Vo0SLNnj3b5e/yfP6%2BZy6GgOUDNm/erPHjx2vWrFkaM2aMJCk8PLxMUnc4HGrUqJECAwPVuHFj5zOn8%2BfDw8Odx7ibb9KkicLDw5Wfn6%2BSkhKXOUlq0qSJJ%2B5ilXz55Zfavn27yyn7c9zd75ycHIWHh5c7f64v5465cP7kyZPOvri7bUBAgPO23nTuUvL5z/qioqJkjFFRUZHf9kU6u2eOHDmixx9/XKGhoYqMjFRiYqI%2B/vhjt/8W/Kk353jyMaM6j1e%2BorS0VDNmzNDmzZv11ltv6frrr3fOeXKPVPSY5k0pKSkaMmSIy2W%2Bc9gzFSNgedm3336rpKQkLVq0SEOGDHGOx8TEaO/evSouLnaOZWZm6sYbb3TOX3iN/tx8cHCwWrZsqV27djnncnNzdejQIXXs2FFt27aVMUY//PCDy23DwsLUvHlzT93VSluzZo2OHz%2Buvn37Ki4uznlqOi4uTq1atSpzv3fu3OnSl/Pvd0lJiXbv3q0bb7xR0dHRatiwocv8jz/%2BqDNnzigmJkYxMTHKyspyvrVfOtuXFi1aqEGDBp68y5XStGlThYSEaM%2BePc6xo0ePqk6dOurdu7ff9kU6e39KS0tdnvGeOXNGktS9e3e/7s05nnzMqM7jla%2BYN2%2BefvrpJ7311luKjo52mfPkHunQocNF96c3rVmzRqtWrVJcXJzi4uL0yiuv6IMPPlBcXBx7pjI8/jJ6lKuoqMjccccd5u233y4zd/r0adO3b1/z/PPPm99//9189913pmvXruaTTz4xxhizd%2B9e06FDB/PJJ5%2BYwsJC8%2B6775rOnTub7OxsY8zZd6n06dPH%2BfbZp59%2B2gwfPty5/pQpU8zDDz9sjh8/brKysszw4cNNSkrKZbnfFXE4HCYrK8v5Z/v27aZVq1YmKyvLHD161HTu3Nm88847prCw0Hz66aemY8eOzneqbNmyxXTp0sVs377d/P7772bx4sWmd%2B/e5tSpU8aYs%2B9SGTp0qPnll1/MiRMnzIQJE8zkyZOdP3vkyJFm1qxZJi8vz%2Bzbt8/069fPrFy50it9cGfevHmmf//%2B5ueffza//fabiY%2BPNzNmzDC//fabX/flxIkTplu3buavf/2r%2Bf33382JEyfMxIkTzb/927/5ZW/cvSPMk48Z1X28upzc9Wbbtm0mNjbW5R3d5/PkHqlof14u7vpy/uNwVlaWmTdvnklMTHR%2BVIK/7JlLRcDyoq%2B//tq0atXKxMTElPlz5MgRs3fvXjNq1CgTExNj%2BvTpY9544w2X269fv94MHDjQtG/f3txzzz3mq6%2B%2Bcs6VlpaaRYsWmVtuucV07NjRPPLIIy6fH5Kbm2umTp1qOnXqZGJjY82cOXPM6dOnL9t9r4rDhw87P6bBGGO%2B%2Buorc/fdd5v27dubgQMHlvnYgjfeeMP07t3bxMTEmNGjR5u9e/c6506fPm1mz55tYmNjTefOnc3jjz9ucnNznfNZWVnm4YcfNh07djTdu3c3zz//vPMzXXzB%2BfV36tTJJCUlmfz8fGOMf/fFGGMyMzPNfffdZ7p27Wq6d%2B9upkyZYo4dO2aM8Z/enHv8aNOmjWnTpo3z%2B3M8%2BZhRncery%2BFivZk5c6bL2Lk/Dz74oPP2ntwjFe1PT6poz5zv%2Beefd/mYhtq%2BZ6orwJga/ioyAAAAH8NrsAAAACxGwAIAALAYAQsAAMBiBCwAAACLEbAAAAAsRsACAACwGAELAADAYgQsAAAAixGwAAAALEbAAgAAsBgBCwAAwGIELAAAAIsRsAAAACz2/wBgR%2BvQhUrPnAAAAABJRU5ErkJggg%3D%3D\"/>\n",
       "        </div>\n",
       "        <div role=\"tabpanel\" class=\"tab-pane col-md-12\" id=\"common-4137530268564546891\">\n",
       "            \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">33195</td>\n",
       "        <td class=\"number\">2</td>\n",
       "        <td class=\"number\">0.4%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">33401</td>\n",
       "        <td class=\"number\">2</td>\n",
       "        <td class=\"number\">0.4%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">23680</td>\n",
       "        <td class=\"number\">2</td>\n",
       "        <td class=\"number\">0.4%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">28915</td>\n",
       "        <td class=\"number\">2</td>\n",
       "        <td class=\"number\">0.4%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">47396</td>\n",
       "        <td class=\"number\">2</td>\n",
       "        <td class=\"number\">0.4%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">38242</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">62823</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">26981</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">32100</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">42339</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"other\">\n",
       "        <td class=\"fillremaining\">Other values (520)</td>\n",
       "        <td class=\"number\">520</td>\n",
       "        <td class=\"number\">97.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "        </div>\n",
       "        <div role=\"tabpanel\" class=\"tab-pane col-md-12\"  id=\"extreme-4137530268564546891\">\n",
       "            <p class=\"h4\">Minimum 5 values</p>\n",
       "            \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">19608</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">21913</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">22893</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">23037</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">23414</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "            <p class=\"h4\">Maximum 5 values</p>\n",
       "            \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">99447</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">108023</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">108608</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">122871</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">143652</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "        </div>\n",
       "    </div>\n",
       "</div>\n",
       "</div><div class=\"row variablerow\">\n",
       "    <div class=\"col-md-3 namecol\">\n",
       "        <p class=\"h4 pp-anchor\" id=\"pp_var_text\">text<br/>\n",
       "            <small>Categorical</small>\n",
       "        </p>\n",
       "    </div><div class=\"col-md-3\">\n",
       "    <table class=\"stats \">\n",
       "        <tr class=\"\">\n",
       "            <th>Distinct count</th>\n",
       "            <td>10</td>\n",
       "        </tr>\n",
       "        <tr>\n",
       "            <th>Unique (%)</th>\n",
       "            <td>1.9%</td>\n",
       "        </tr>\n",
       "        <tr class=\"ignore\">\n",
       "            <th>Missing (%)</th>\n",
       "            <td>0.0%</td>\n",
       "        </tr>\n",
       "        <tr class=\"ignore\">\n",
       "            <th>Missing (n)</th>\n",
       "            <td>0</td>\n",
       "        </tr>\n",
       "    </table>\n",
       "</div>\n",
       "<div class=\"col-md-6 collapse in\" id=\"minifreqtable-3460658111431752074\">\n",
       "    <table class=\"mini freq\">\n",
       "        <tr class=\"\">\n",
       "    <th>a05</th>\n",
       "    <td>\n",
       "        <div class=\"bar\" style=\"width:18%\" data-toggle=\"tooltip\" data-placement=\"right\" data-html=\"true\"\n",
       "             data-delay=500 title=\"Percentage: 11.6%\">\n",
       "            &nbsp;\n",
       "        </div>\n",
       "        62\n",
       "    </td>\n",
       "</tr><tr class=\"\">\n",
       "    <th>a02</th>\n",
       "    <td>\n",
       "        <div class=\"bar\" style=\"width:17%\" data-toggle=\"tooltip\" data-placement=\"right\" data-html=\"true\"\n",
       "             data-delay=500 title=\"Percentage: 11.2%\">\n",
       "            &nbsp;\n",
       "        </div>\n",
       "        60\n",
       "    </td>\n",
       "</tr><tr class=\"\">\n",
       "    <th>a07</th>\n",
       "    <td>\n",
       "        <div class=\"bar\" style=\"width:16%\" data-toggle=\"tooltip\" data-placement=\"right\" data-html=\"true\"\n",
       "             data-delay=500 title=\"Percentage: 10.3%\">\n",
       "            &nbsp;\n",
       "        </div>\n",
       "        55\n",
       "    </td>\n",
       "</tr><tr class=\"other\">\n",
       "    <th>Other values (7)</th>\n",
       "    <td>\n",
       "        <div class=\"bar\" style=\"width:100%\" data-toggle=\"tooltip\" data-placement=\"right\" data-html=\"true\"\n",
       "             data-delay=500 title=\"Percentage: 66.9%\">\n",
       "            358\n",
       "        </div>\n",
       "        \n",
       "    </td>\n",
       "</tr>\n",
       "    </table>\n",
       "</div>\n",
       "<div class=\"col-md-12 text-right\">\n",
       "    <a role=\"button\" data-toggle=\"collapse\" data-target=\"#freqtable-3460658111431752074, #minifreqtable-3460658111431752074\"\n",
       "       aria-expanded=\"true\" aria-controls=\"collapseExample\">\n",
       "        Toggle details\n",
       "    </a>\n",
       "</div>\n",
       "<div class=\"col-md-12 extrapadding collapse\" id=\"freqtable-3460658111431752074\">\n",
       "    \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">a05</td>\n",
       "        <td class=\"number\">62</td>\n",
       "        <td class=\"number\">11.6%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">a02</td>\n",
       "        <td class=\"number\">60</td>\n",
       "        <td class=\"number\">11.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:96%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">a07</td>\n",
       "        <td class=\"number\">55</td>\n",
       "        <td class=\"number\">10.3%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:88%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">b10</td>\n",
       "        <td class=\"number\">52</td>\n",
       "        <td class=\"number\">9.7%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:84%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">b09</td>\n",
       "        <td class=\"number\">52</td>\n",
       "        <td class=\"number\">9.7%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:84%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">b03</td>\n",
       "        <td class=\"number\">52</td>\n",
       "        <td class=\"number\">9.7%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:84%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">b02</td>\n",
       "        <td class=\"number\">51</td>\n",
       "        <td class=\"number\">9.5%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:82%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">a04</td>\n",
       "        <td class=\"number\">51</td>\n",
       "        <td class=\"number\">9.5%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:82%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">b01</td>\n",
       "        <td class=\"number\">51</td>\n",
       "        <td class=\"number\">9.5%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:82%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">a01</td>\n",
       "        <td class=\"number\">49</td>\n",
       "        <td class=\"number\">9.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:79%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "</div>\n",
       "</div><div class=\"row variablerow ignore\">\n",
       "    <div class=\"col-md-3 namecol\">\n",
       "        <p class=\"h4 pp-anchor\" id=\"pp_var_x\"><s>x</s><br/>\n",
       "            <small>Unsupported</small>\n",
       "        </p>\n",
       "    </div><div class=\"col-md-3\">\n",
       "    <p><em>This variable is an unsupported type, check if it needs cleaning or further analysis</em></p>\n",
       "</div>\n",
       "<div class=\"col-md-6\">\n",
       "    <table class=\"stats \">\n",
       "        <tr>\n",
       "            <th>Unsupported value</th>\n",
       "        </tr>\n",
       "    </table>\n",
       "</div>\n",
       "</div><div class=\"row variablerow\">\n",
       "    <div class=\"col-md-3 namecol\">\n",
       "        <p class=\"h4 pp-anchor\" id=\"pp_var_y\">y<br/>\n",
       "            <small>Numeric</small>\n",
       "        </p>\n",
       "    </div><div class=\"col-md-6\">\n",
       "    <div class=\"row\">\n",
       "        <div class=\"col-sm-6\">\n",
       "            <table class=\"stats \">\n",
       "                <tr>\n",
       "                    <th>Distinct count</th>\n",
       "                    <td>7</td>\n",
       "                </tr>\n",
       "                <tr>\n",
       "                    <th>Unique (%)</th>\n",
       "                    <td>1.3%</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Missing (%)</th>\n",
       "                    <td>0.0%</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Missing (n)</th>\n",
       "                    <td>0</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Infinite (%)</th>\n",
       "                    <td>0.0%</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Infinite (n)</th>\n",
       "                    <td>0</td>\n",
       "                </tr>\n",
       "            </table>\n",
       "\n",
       "        </div>\n",
       "        <div class=\"col-sm-6\">\n",
       "            <table class=\"stats \">\n",
       "\n",
       "                <tr>\n",
       "                    <th>Mean</th>\n",
       "                    <td>2.7065</td>\n",
       "                </tr>\n",
       "                <tr>\n",
       "                    <th>Minimum</th>\n",
       "                    <td>0</td>\n",
       "                </tr>\n",
       "                <tr>\n",
       "                    <th>Maximum</th>\n",
       "                    <td>6</td>\n",
       "                </tr>\n",
       "                <tr class=\"alert\">\n",
       "                    <th>Zeros (%)</th>\n",
       "                    <td>23.7%</td>\n",
       "                </tr>\n",
       "            </table>\n",
       "        </div>\n",
       "    </div>\n",
       "</div>\n",
       "<div class=\"col-md-3 collapse in\" id=\"minihistogram-3389567407320575927\">\n",
       "    <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAMgAAABLCAYAAAA1fMjoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD%2BnaQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAASNJREFUeJzt28EJwkAQQFEVS7IIe/JsTxZhT%2Btd5GOEkCW%2Bdw/s5WeGhD2OMcYB%2BOi09QFgZuetD/DucnssfuZ5v65wEjBBIAkEgkAgCASCQCAIBIJAIAgEgkAgCASCQCAIBIJAIAgEgkAgCATCdBemfuGSFWvZRSDMaQ8vLisWBIFAsGLtwNJVZrY1ZmYCmcwvezvrsWJBMEH4yr9ONhMEgkAgCASCQCAIBIKvWAv4Ifd/TBAIAoEgEAgCgSAQCAKBIBAIAoEgEAgCgSAQCAKBIBAIAoEgEAgCgXAcY4ytDwGzMkEgCASCQCAIBIJAIAgEgkAgCASCQCAIBIJAIAgEgkAgCASCQCAIBIJAIAgEgkAgCASCQCAIBIJAIAgEgkAgvACLkxppDzMr7QAAAABJRU5ErkJggg%3D%3D\">\n",
       "\n",
       "</div>\n",
       "<div class=\"col-md-12 text-right\">\n",
       "    <a role=\"button\" data-toggle=\"collapse\" data-target=\"#descriptives-3389567407320575927,#minihistogram-3389567407320575927\"\n",
       "       aria-expanded=\"false\" aria-controls=\"collapseExample\">\n",
       "        Toggle details\n",
       "    </a>\n",
       "</div>\n",
       "<div class=\"row collapse col-md-12\" id=\"descriptives-3389567407320575927\">\n",
       "    <ul class=\"nav nav-tabs\" role=\"tablist\">\n",
       "        <li role=\"presentation\" class=\"active\"><a href=\"#quantiles-3389567407320575927\"\n",
       "                                                  aria-controls=\"quantiles-3389567407320575927\" role=\"tab\"\n",
       "                                                  data-toggle=\"tab\">Statistics</a></li>\n",
       "        <li role=\"presentation\"><a href=\"#histogram-3389567407320575927\" aria-controls=\"histogram-3389567407320575927\"\n",
       "                                   role=\"tab\" data-toggle=\"tab\">Histogram</a></li>\n",
       "        <li role=\"presentation\"><a href=\"#common-3389567407320575927\" aria-controls=\"common-3389567407320575927\"\n",
       "                                   role=\"tab\" data-toggle=\"tab\">Common Values</a></li>\n",
       "        <li role=\"presentation\"><a href=\"#extreme-3389567407320575927\" aria-controls=\"extreme-3389567407320575927\"\n",
       "                                   role=\"tab\" data-toggle=\"tab\">Extreme Values</a></li>\n",
       "\n",
       "    </ul>\n",
       "\n",
       "    <div class=\"tab-content\">\n",
       "        <div role=\"tabpanel\" class=\"tab-pane active row\" id=\"quantiles-3389567407320575927\">\n",
       "            <div class=\"col-md-4 col-md-offset-1\">\n",
       "                <p class=\"h4\">Quantile statistics</p>\n",
       "                <table class=\"stats indent\">\n",
       "                    <tr>\n",
       "                        <th>Minimum</th>\n",
       "                        <td>0</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>5-th percentile</th>\n",
       "                        <td>0</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Q1</th>\n",
       "                        <td>1</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Median</th>\n",
       "                        <td>3</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Q3</th>\n",
       "                        <td>5</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>95-th percentile</th>\n",
       "                        <td>6</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Maximum</th>\n",
       "                        <td>6</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Range</th>\n",
       "                        <td>6</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Interquartile range</th>\n",
       "                        <td>4</td>\n",
       "                    </tr>\n",
       "                </table>\n",
       "            </div>\n",
       "            <div class=\"col-md-4 col-md-offset-2\">\n",
       "                <p class=\"h4\">Descriptive statistics</p>\n",
       "                <table class=\"stats indent\">\n",
       "                    <tr>\n",
       "                        <th>Standard deviation</th>\n",
       "                        <td>2.1624</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Coef of variation</th>\n",
       "                        <td>0.79895</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Kurtosis</th>\n",
       "                        <td>-1.3952</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Mean</th>\n",
       "                        <td>2.7065</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>MAD</th>\n",
       "                        <td>1.9232</td>\n",
       "                    </tr>\n",
       "                    <tr class=\"\">\n",
       "                        <th>Skewness</th>\n",
       "                        <td>0.14663</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Sum</th>\n",
       "                        <td>1448</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Variance</th>\n",
       "                        <td>4.6759</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Memory size</th>\n",
       "                        <td>4.3 KiB</td>\n",
       "                    </tr>\n",
       "                </table>\n",
       "            </div>\n",
       "        </div>\n",
       "        <div role=\"tabpanel\" class=\"tab-pane col-md-8 col-md-offset-2\" id=\"histogram-3389567407320575927\">\n",
       "            <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAlgAAAGQCAYAAAByNR6YAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD%2BnaQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3X1clHW%2B//E3MEIBglCBe8jC1rIQVjOVjcwburH0aN5g6Dl7jNOdFkmSbJo3D3XXDXuoWx3xmLarZfXY5uDpBGKax7XcbmjPsduBzGOu7a48MmaRCTVMYfj90U%2B2iWJQvnBdM/N6Ph7%2Bwfearus9H2bs7XVdDGEtLS0tAgAAgDHhVgcAAAAINhQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGCYw%2BoAocLtPmZ8n%2BHhYUpMjNHRoyfk9bYY338wYEbtYz7%2BMSP/mFH7mI9/XTmjiy7qaXR/HcUZrAAWHh6msLAwhYeHWR3FtphR%2B5iPf8zIP2bUPubjXzDOiIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIY5rA6AzhmycIfVETps%2B5zrrI4AAEC34AwWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAwL2oL1xhtvKCsrS4WFhW227dy5UxMmTNDVV1%2BtMWPG6D/%2B4z98tm/evFljxozR4MGDNX36dFVVVXVXbAAAEAQcVgfoCk8//bS2bNmiSy%2B9tM22jz76SEVFRfr1r3%2BtUaNG6a233lJ%2Bfr4uu%2BwyDRkyRLt379aaNWv0m9/8Rv3799fmzZs1a9Ys7dy5U9HR0RY8GwAAEGiC8gxWVFTUDxYsj8ejmTNn6sYbb5TD4dDIkSN1xRVXaO/evZIkp9OpyZMna%2BDAgTrvvPN09913S5Jee%2B21bn0OAAAgcAXlGawZM2b84LYRI0ZoxIgRrV83NTXJ7XYrOTlZklRdXa2xY8e2bg8PD9dVV10ll8ulcePGdej4tbW1crvdPmsOR7SSkpLO5mn4FRERWP3Y4ej%2BvGdmFGiz6i7Mxz9m5B8zah/z8S8YZxSUBetsrFq1StHR0a2lyuPxKD4%2B3ucx8fHxqq%2Bv7/A%2BnU6nSkpKfNby8/NVUFDQ%2BcABLCEhxrJjx8Wdb9mxAwHz8Y8Z%2BceM2sd8/AumGYVswWppadGqVatUUVGhzZs3KyoqymdbZ%2BTm5io7O9tnzeGIVn39iU7t97sCrembfv4dERERrri489XQ0KjmZm%2B3H9/umI9/zMg/ZtQ%2B5uNfV87Iqn/ch2TB8nq9euSRR/TRRx/pd7/7nfr06dO6LSEhQR6Px%2BfxHo9Hl19%2BeYf3n5SU1OZyoNt9TE1Nof3GsvL5Nzd7Q37%2B7WE%2B/jEj/5hR%2B5iPf8E0o8A6BWLIo48%2BqgMHDrQpV5KUnp6u6urq1q%2Bbm5v18ccfa%2BDAgd0dEwAABKiQK1jvvvuuysvLtWHDBvXq1avN9unTp%2Bvll1/WBx98oMbGRq1bt06RkZEaNWpU94cFAAABKSgvEWZkZEj65icEJWnXrl2SJJfLpf/8z//UsWPHNHr0aJ//ZujQodq4caNGjBihhx56SHPmzFFdXZ0yMjK0YcMGnXfeed37JAAAQMAKa%2BnsHd3oELf7mPF9OhzhumnVG8b321W2z7mu24/pcIQrISFG9fUngua6vknMxz9m5B8zah/z8a8rZ3TRRT2N7q%2BjQu4SIQAAQFejYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAULAADAsKAtWG%2B88YaysrJUWFjYZtsrr7yi8ePH6%2Bqrr9bkyZP15ptvtm7zer16/PHHdcMNN2jo0KG666679Ne//rU7owMAgAAXlAXr6aef1vLly3XppZe22bZv3z7NmzdPRUVFeuedd5SXl6cHHnhAR44ckSS98MIL2rp1qzZs2KDXXntNqampys/PV0tLS3c/DQAAEKCCsmBFRUVpy5Yt31uwSktLNXLkSI0cOVJRUVGaMGGCrrjiCpWXl0uSnE6n8vLy9OMf/1ixsbEqLCzUwYMH9eGHH3b30wAAAAHKYXWArjBjxowf3FZdXa2RI0f6rKWlpcnlcunkyZP69NNPlZaW1rotNjZWl156qVwulwYNGtSh49fW1srtdvusORzRSkpKOotn4V9ERGD1Y4ej%2B/OemVGgzaq7MB//mJF/zKh9zMe/YJxRUBas9ng8HsXHx/usxcfH69NPP9WXX36plpaW791eX1/f4WM4nU6VlJT4rOXn56ugoODcgweBhIQYy44dF3e%2BZccOBMzHP2bkHzNqH/PxL5hmFHIFS5Lf%2B6k6e79Vbm6usrOzfdYcjmjV15/o1H6/K9Cavunn3xEREeGKiztfDQ2Nam72dvvx7Y75%2BMeM/GNG7WM%2B/nXljKz6x33IFayEhAR5PB6fNY/Ho8TERPXq1Uvh4eHfu/2CCy7o8DGSkpLaXA50u4%2BpqSm031hWPv/mZm/Iz789zMc/ZuQfM2of8/EvmGYUWKdADEhPT1dVVZXPmsvl0sCBAxUVFaXLL79c1dXVrdsaGhr0l7/8RT/5yU%2B6OyoAAAhQIVewbr/9dr399tt6/fXX9fXXX2vLli367LPPNGHCBEnS9OnTtXnzZh08eFDHjx/XqlWrdNVVVykjI8Pi5AAAIFAE5SXCM2WoqalJkrRr1y5J35ypuuKKK7Rq1SoVFxerpqZG/fr10/r163XRRRdJkqZNmya3261/%2BZd/0YkTJ5SZmdnmhnUAAID2BGXBcrlc7W6/%2BeabdfPNN3/vtrCwMBUUFIT8T/wBAIBzF3KXCAEAALoaBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhoVswfr44481Y8YMDRkyRNddd52Kiop09OhRSVJlZaVycnI0ePBgjRs3TuXl5RanBQAAgSQkC1ZTU5PuvfdeDRo0SG%2B//bYqKip09OhRLV26VLW1tbr//vs1bdo0VVZWauHChVq8eLFcLpfVsQEAQIBwWB3ACm63W263W7fddpsiIyMVGRmpm266SRs3btTWrVuVmpqqnJwcSVJWVpays7NVWlqqjIwMi5MHtlufeMvqCGdl%2B5zrrI4AAAhQtitY2dnZmjx5sqZMmaIf/ehHXXKM5ORkXXXVVXI6nXrwwQd18uRJ7dy5U6NGjVJ1dbXS0tJ8Hp%2BWlqbt27d3eP%2B1tbVyu90%2Baw5HtJKSkozkPyMiIiRPQHYbhyP453vmNcRr6YcxI/%2BYUfuYj3/BOCPbFawpU6Zo27ZtWrduna699lrdfvvtys7OlsNhLmp4eLjWrFmjvLw8Pfvss5KkYcOGae7cubr//vuVnJzs8/hevXqpvr6%2Bw/t3Op0qKSnxWcvPz1dBQUHnw6PbJCTEWB2h28TFnW91BNtjRv4xo/YxH/%2BCaUZhLS0tLVaH%2BD7V1dWqqKjQ9u3bdfr0aU2cOFE5OTnq27dvp/d96tQpTZo0SaNGjdKsWbP01VdfadmyZQoPD9dXX32l/v37a968ea2PLy0t1b//%2B7/rtdde69D%2Bu/MMVvZje4zuE3/330XXWx2hy0VEhCsu7nw1NDSqudlrdRxbYkb%2BMaP2MR//unJGVv1j2XZnsM4YMGCABgwYoIcfflivvPKKli5dqo0bNyorK0sPPvigfvKTn5zzvisrK3X48GE99NBDioiIUM%2BePVVQUKDbbrtN119/vTwej8/j6%2BvrlZiY2OH9JyUltSlTbvcxNTXxxgokofT9am72htTzPRfMyD9m1D4r5hNI977u/dUtQfUasu3FztOnT%2BuVV17RPffco3nz5ik5OVmPPPKIrrrqKuXl5Wnr1q3nvO/m5mZ5vV59%2B%2BTdqVOnJH1zU3tVVZXP46uqqjRw4MBzPh4AAAgttjuDdfDgQW3ZskUvv/yyTpw4oTFjxujZZ5/VNddc0/qYoUOHaunSpRo/fvw5HePqq69WdHS01qxZo1mzZunkyZNat26dhg4dqttuu00lJSUqLS3VhAkT9M4772jPnj1yOp2mniIAAAhytitY48aNU9%2B%2BfTVz5kxNnDhRvXr1avOYkSNHtn4o6LlISEjQb3/7Wz322GMaMWKEIiMjNWzYMC1dulQXXHCB1q9fr%2BXLl2vZsmVKSUnRypUrdeWVV3bmaQEAgBBiu4K1efNmDRs2zO/jPvzww04dJz09Xc8999z3bhs6dKjKyso6tX8AABC6bHcPVv/%2B/TVr1izt2rWrde2ZZ57RPffc0%2BbmcwAAADuyXcEqLi7WsWPH1K9fv9a1UaNGyev1asWKFRYmAwAA6BjbXSJ88803tXXrViUkJLSupaamatWqVfrHf/xHC5MBAAB0jO3OYJ08eVJRUVFt1sPDw9XY2GhBIgAAgLNju4I1dOhQrVixQl9%2B%2BWXr2hdffKFly5b5fFQDAACAXdnuEuGCBQt055136tprr1VsbKy8Xq9OnDihPn36/OBP/QEAANiJ7QpWnz59tG3bNv3hD3/QX/7yF4WHh6tv374aPny4IiIirI4HAADgl%2B0KliRFRkbqxhtvtDoGAADAObFdwfrrX/%2Bq1atX68CBAzp58mSb7b///e8tSAUAANBxtitYCxYsUG1trYYPH67o6Gir4wAAAJw12xWsqqoq/f73v1diYqLVUQAAAM6J7T6m4YILLuDMFQAACGi2K1gzZ85USUmJWlparI4CAABwTmx3ifAPf/iD3nvvPb300ku6%2BOKLFR7u2wFffPFFi5IBAAB0jO0KVmxsrEaMGGF1DAAAgHNmu4JVXFxsdQQACGi3PvGW1RE6bPuc66yOAHQJ292DJUl/%2BtOftGbNGj3yyCOta%2B%2B//76FiQAAADrOdgWrsrJSEyZM0M6dO1VRUSHpmw8fnTFjBh8yCgAAAoLtCtbjjz%2Bun//859q6davCwsIkffP7CVesWKG1a9danA4AAMA/292D9X//9396/vnnJam1YEnSLbfcogULFlgVC0AIG7Jwh9URAAQY253B6tmz5/f%2BDsLa2lpFRkZakAgAAODs2K5gDR48WI8%2B%2BqiOHz/eunbo0CHNmzdP1157rYXJAAAAOsZ2lwgfeeQR3XHHHcrMzFRzc7MGDx6sxsZGXX755VqxYoXV8QAAAPyyXcHq3bu3KioqtGfPHh06dEjnnXee%2Bvbtq%2Buuu87nniwAAAC7sl3BkqQePXroxhtvtDoGAADAObFdwcrOzm73TBWfhQUAAOzOdgVr7NixPgWrublZhw4dksvl0h133GFhMgAAgI6xXcEqKir63vVXX31Vf/zjH7s5DQAAwNmz3cc0/JAbb7xR27ZtszoGAACAXwFTsD7%2B%2BGO1tLRYHQMAAMAv210inDZtWpu1xsZGHTx4UDfffLMFiQAAAM6O7QpWampqm58ijIqKUk5OjqZOnWpRKgAAgI6zXcHi09oBAECgs13Bevnllzv82IkTJ3ZhEgAAgHNju4K1cOFCeb3eNje0h4WF%2BayFhYVRsAAAgC3ZrmD95je/0caNGzVr1iz1799fLS0t2r9/v55%2B%2Bmn97Gc/U2ZmptURAQAA2mW7grVixQpt2LBBycnJrWtDhgxRnz59dNddd6miosLCdACAUHbrE29ZHQEBwnafg/XZZ58pPj6%2BzXpcXJxqamqMHmvdunUaPny4Bg0apLy8PB0%2BfFiSVFlZqZycHA0ePFjjxo1TeXm50eMCAIDgZruClZKSohUrVqi%2Bvr51raGhQatXr9Yll1xi7DgvvPCCysvLtXnzZr355pvq16%2BfnnnmGdXW1ur%2B%2B%2B/XtGnTVFlZqYULF2rx4sVyuVzGjg0AAIKb7S4RLliwQHPnzpXT6VRMTIzCw8N1/PhxnXfeeVq7dq2x42zcuFHz5s3TZZddJklatGiRJOm3v/2tUlNTlZOTI0nKyspSdna2SktLlZGRYez4AAAgeNmuYA0fPlyvv/669uzZoyNHjqilpUXJycm6/vrr1bNnTyPH%2BOKLL3T48GF9%2BeWXGjt2rOrq6pSZmamlS5equrpaaWlpPo9PS0vT9u3bO7z/2tpaud1unzWHI1pJSUlG8p8REWG7E5BBxeEI/vmeeQ3xWvphzKZrhcL7DB0XTO832xUsSTr//PN1ww036MiRI%2BrTp4/x/R85ckSStGPHDm3atEktLS0qKCjQokWLdPLkSZ8b7CWpV69ePpcs/XE6nSopKfFZy8/PV0FBQefDo9skJMRYHaHbxMWdb3UEhKhQep/Bv2D6u8h2BevkyZNasmSJtm3bJkmqqqpSQ0ODHnroIf36179WXFxcp49x5vO07r777tYyNXv2bN1zzz3Kysrq9P5zc3OVnZ3ts%2BZwRKu%2B/kSn9/1twdT07cj098uOIiLCFRd3vhoaGtXc7LU6ji3xPutaofA%2BQ8d1xd9FVpV42xWslStXat%2B%2BfVq1apUefvjh1vXm5matWrVKv/jFLzp9jAsvvFCSfMpaSkqKWlpadPr0aXk8Hp/H19fXKzExscP7T0pKanM50O0%2BpqYm/gcWSELp%2B9Xc7A2p5wv74HWHbwumv4ts90%2BzV199Vf/2b/%2BmW265pfWXPsfFxam4uFg7d%2B40cozevXsrNjZW%2B/bta12rqalRjx49NHLkSFVVVfk8vqqqSgMHDjRybAAAEPxsV7BOnDih1NTUNuuJiYn66quvjBzD4XAoJydHTz31lP785z%2Brrq5Oa9eu1fjx4zVp0iTV1NSotLRUX3/9tfbs2aM9e/bo9ttvN3JsAAAQ/Gx3ifCSSy7RH//4R2VmZvr87sEdO3boH/7hH4wdZ%2B7cuTp16pSmTp2q06dPa8yYMVq0aJFiYmK0fv16LV%2B%2BXMuWLVNKSopWrlypK6%2B80tixAQBAcLNdwfqnf/onzZ49W1OmTJHX69WmTZtUVVWlV199VQsXLjR2nMjISC1ZskRLlixps23o0KEqKyszdiwAABBabFewcnNz5XA49PzzzysiIkJPPfWU%2Bvbtq1WrVumWW26xOh4AAIBftitYR48e1ZQpUzRlyhSrowAAAJwT293kfsMNN/jcewUAABBobFewMjMzz%2BrX0gAAANiN7S4R/uhHP9KvfvUrbdiwQZdccol69Ojhs3316tUWJQMAAOgY2xWsTz/9VJdddpkkndXv/wMAALAL2xSswsJCPf7443ruueda19auXav8/HwLUwEAAJw929yDtXv37jZrGzZssCAJAABA59imYH3fTw7y04QAACAQ2aZgnfnFzv7WAAAA7M42BQsAACBYULAAAAAMs81PEZ4%2BfVpz5871u8bnYAEAALuzTcG65pprVFtb63cNAADA7mxTsL79%2BVcAAACBjHuwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGCYbX7ZM4DOufWJt6yO0GHb51xndQQA6FKcwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwLOQL1qOPPqr%2B/fu3fl1ZWamcnBwNHjxY48aNU3l5uYXpAABAIArpX5Wzb98%2BlZWVtX5dW1ur%2B%2B%2B/XwsXLtT48eP17rvv6r777lPfvn2VkZFhYVIAABBIQvYMltfr1ZIlS5SXl9e6tnXrVqWmpionJ0dRUVHKyspSdna2SktLrQsKAAACTsiewXrxxRcVFRWl8ePH64knnpAkVVdXKy0tzedxaWlp2r59%2B1ntu7a2Vm6322fN4YhWUlJS50J/R0REyPbjbuFwMN%2BuEkiz5X3WtQLptYCuF0zvt5AsWH/729%2B0Zs0aPffccz7rHo9HycnJPmu9evVSfX39We3f6XSqpKTEZy0/P18FBQXnFhiWSEiIsTpC0GK2OIPXAr4tLu58qyMYE5IFq7i4WJMnT1a/fv10%2BPBh4/vPzc1Vdna2z5rDEa36%2BhNGjxNMTd%2BOTH%2B/8HeBNFveZ10rkF4L6HoNDY1qbvYa3adVJT7kClZlZaXef/99VVRUtNmWkJAgj8fjs1ZfX6/ExMSzOkZSUlKby4Fu9zE1NZl90aBr8f3qOswWZ/BawLc1N3uD5jURcgWrvLxcdXV1Gj16tCSppaVFkpSZmak777yzTfGqqqrSwIEDuz0nAAAIXCFXsObPn68HH3yw9esjR44oNzdXZWVl8nq9Wr9%2BvUpLSzVhwgS988472rNnj5xOp4WJAQBAoAm5ghUfH6/4%2BPjWr5uamiRJvXv3liStX79ey5cv17Jly5SSkqKVK1fqyiuvtCQrAAAITCFXsL7r4osv1v79%2B1u/Hjp0qM%2BHjwIAAJwtfjwGAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAwL2YJVU1Oj/Px8ZWZmKisrS/Pnz1dDQ4Mkad%2B%2BffrZz36ma665RjfffLM2btxocVoAABBIQrZgzZo1S3Fxcdq9e7deeuklHThwQI899phOnjypmTNn6qc//aneeOMNPf7441q/fr127txpdWQAABAgQrJgNTQ0KD09XXPnzlVMTIx69%2B6tSZMmae/evXr99dd1%2BvRp3XfffYqOjtaAAQM0depUOZ1Oq2MDAIAA4bA6gBXi4uJUXFzss/b5558rKSlJ1dXV6t%2B/vyIiIlq3paWlqbS0tMP7r62tldvt9llzOKKVlJTUueDfERERkv242zgczLerBNJseZ91rUB6LaDrBdP7LSQL1ne5XC49//zzWrdunbZv3664uDif7b169ZLH45HX61V4uP9vvtPpVElJic9afn6%2BCgoKjOZG10pIiLE6QtBitjiD1wK%2BLS7ufKsjGBPyBevdd9/Vfffdp7lz5yorK0vbt2//3seFhYV1eJ%2B5ubnKzs72WXM4olVff6JTWb8rmJq%2BHZn%2BfuHvAmm2vM%2B6ViC9FtD1Ghoa1dzsNbpPq0p8SBes3bt36%2Bc//7kWL16siRMnSpISExP12Wef%2BTzO4/GoV69eHTp7JUlJSUltLge63cfU1GT2RYOuxfer6zBbnMFrAd/W3OwNmtdEyP7T7L333tO8efP05JNPtpYrSUpPT9f%2B/fvV1NTUuuZyuTRw4EArYgIAgAAUkgWrqalJixYtUlFRkYYPH%2B6zbeTIkYqNjdW6devU2NioDz/8UFu2bNH06dMtSgsAAAJNSBasDz74QAcPHtTy5cuVkZHh88ftduupp57S22%2B/rWHDhmnOnDkqLCzUqFGjrI4NAAACREjegzVkyBDt37%2B/3cf87ne/66Y0AAAg2ITkGSwAAICuRMECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEUrO9RU1Oje%2B%2B9V5mZmRo9erRWrlwpr9drdSwAABAgHFYHsKPZs2drwIAB2rVrl%2Brq6jRz5kxdeOGF%2Btd//VerowEAgADAGazvcLlc%2BuSTT1RUVKSePXsqNTVVeXl5cjqdVkcDAAABgjNY31FdXa2UlBTFx8e3rg0YMECHDh3S8ePHFRsb63cftbW1crvdPmsOR7SSkpKMZo2IoB93JYeD%2BXaVQJot77OuFUivBXS9YHq/UbC%2Bw%2BPxKC4uzmftTNmqr6/vUMFyOp0qKSnxWXvggQc0e/Zsc0H1TZG7o/cB5ebmGi9vwaK2tlZOpzMkZrT3V7ec9X8TSvM5V7zP/Aul1xHvs65RW1urNWvWBNWMgqcqGtTS0tKp/z43N1cvvfSSz5/c3FxD6f7O7XarpKSkzdky/B0zah9YPH/jAAAG9UlEQVTz8Y8Z%2BceM2sd8/AvGGXEG6zsSExPl8Xh81jwej8LCwpSYmNihfSQlJQVNAwcAAGePM1jfkZ6ers8//1xHjx5tXXO5XOrXr59iYmIsTAYAAAIFBes70tLSlJGRodWrV%2Bv48eM6ePCgNm3apOnTp1sdDQAABIiIpUuXLrU6hN1cf/31qqio0C9/%2BUtt27ZNOTk5uuuuuxQWFmZ1tDZiYmI0bNgwzq61gxm1j/n4x4z8Y0btYz7%2BBduMwlo6e0c3AAAAfHCJEAAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwClYAqqmp0b333qvMzEyNHj1aK1eulNfrtTqW7bzxxhvKyspSYWGh1VFsqaamRvn5%2BcrMzFRWVpbmz5%2BvhoYGq2PZyieffKI77rhD11xzjbKysjRnzhy53W6rY9nSo48%2Bqv79%2B1sdw3b69%2B%2Bv9PR0ZWRktP755S9/aXUs21m3bp2GDx%2BuQYMGKS8vT4cPH7Y6UqdRsALQ7NmzlZycrF27dmnTpk3atWuXnn32Watj2crTTz%2Bt5cuX69JLL7U6im3NmjVLcXFx2r17t1566SUdOHBAjz32mNWxbOPUqVO68847NWzYMFVWVqqiokJ1dXXi17e2tW/fPpWVlVkdw7Z27Nghl8vV%2Bmfx4sVWR7KVF154QeXl5dq8ebPefPNN9evXT88884zVsTqNghVgXC6XPvnkExUVFalnz55KTU1VXl6enE6n1dFsJSoqSlu2bKFg/YCGhgalp6dr7ty5iomJUe/evTVp0iTt3bvX6mi20djYqMLCQs2cOVORkZFKTEzUTTfdpAMHDlgdzVa8Xq%2BWLFmivLw8q6MgQG3cuFGFhYW67LLLFBsbq0WLFmnRokVWx%2Bo0ClaAqa6uVkpKiuLj41vXBgwYoEOHDun48eMWJrOXGTNmqGfPnlbHsK24uDgVFxfrwgsvbF37/PPPlZSUZGEqe4mPj9fUqVPlcDgkSX/605/0X//1X7r11lstTmYvL774oqKiojR%2B/Hiro9jW6tWrNWrUKA0ZMkSLFy/WiRMnrI5kG1988YUOHz6sL7/8UmPHjlVmZqYKCgp09OhRq6N1GgUrwHg8HsXFxfmsnSlb9fX1VkRCEHC5XHr%2B%2Bed13333WR3FdmpqapSenq6xY8cqIyNDBQUFVkeyjb/97W9as2aNlixZYnUU2xo0aJCysrK0c%2BdOOZ1OffDBB1q2bJnVsWzjyJEjkr65jLpp0yaVlZXpyJEjnMGCNVpaWqyOgCDy7rvv6q677tLcuXOVlZVldRzbSUlJkcvl0o4dO/TZZ5/p4YcftjqSbRQXF2vy5Mnq16%2Bf1VFsy%2Bl0aurUqYqMjNSPf/xjFRUVqaKiQqdOnbI6mi2c%2Bf/Z3XffreTkZPXu3VuzZ8/W7t279fXXX1ucrnMoWAEmMTFRHo/HZ83j8SgsLEyJiYkWpUKg2r17t%2B69914tWLBAM2bMsDqObYWFhSk1NVWFhYWqqKgIissXnVVZWan3339f%2Bfn5VkcJKBdffLGam5tVV1dndRRbOHObwrevzKSkpKilpSXgZ0TBCjDp6en6/PPPff6Cd7lc6tevn2JiYixMhkDz3nvvad68eXryySc1ceJEq%2BPYTmVlpcaMGePzESjh4d/8ldmjRw%2BrYtlGeXm56urqNHr0aGVmZmry5MmSpMzMTG3bts3idPbw8ccfa8WKFT5rBw8eVGRkJPc7/n%2B9e/dWbGys9u3b17pWU1OjHj16BPyMKFgBJi0tTRkZGVq9erWOHz%2BugwcPatOmTZo%2BfbrV0RBAmpqatGjRIhUVFWn48OFWx7Gl9PR0HT9%2BXCtXrlRjY6OOHj2qNWvWaMiQIfwAhaT58%2Bfr1VdfVVlZmcrKyrRhwwZJUllZmbKzsy1OZw8XXHCBnE6nNmzYoFOnTunQoUN68sknlZubq4iICKvj2YLD4VBOTo6eeuop/fnPf1ZdXZ3Wrl2r8ePHt/6ASaAKa%2BGGnoBz5MgRLV68WP/zP/%2Bj2NhYTZs2TQ888IDCwsKsjmYbGRkZkr4pEpJa36gul8uyTHayd%2B9e/fM//7MiIyPbbNuxY4dSUlIsSGU/%2B/fv1/Lly/XRRx8pOjpaP/3pTzV//nwlJydbHc12Dh8%2BrBtuuEH79%2B%2B3Ooqt/O///q9Wr16t/fv3KzIyUpMmTVJhYaGioqKsjmYbp06dUnFxsbZt26bTp09rzJgxWrx4ccBflaFgAQAAGMYlQgAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAw7P8BA2jt9WoyegQAAAAASUVORK5CYII%3D\"/>\n",
       "        </div>\n",
       "        <div role=\"tabpanel\" class=\"tab-pane col-md-12\" id=\"common-3389567407320575927\">\n",
       "            \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">0</td>\n",
       "        <td class=\"number\">127</td>\n",
       "        <td class=\"number\">23.7%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">1</td>\n",
       "        <td class=\"number\">81</td>\n",
       "        <td class=\"number\">15.1%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:64%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">6</td>\n",
       "        <td class=\"number\">79</td>\n",
       "        <td class=\"number\">14.8%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:62%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">4</td>\n",
       "        <td class=\"number\">71</td>\n",
       "        <td class=\"number\">13.3%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:56%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">3</td>\n",
       "        <td class=\"number\">69</td>\n",
       "        <td class=\"number\">12.9%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:54%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">5</td>\n",
       "        <td class=\"number\">62</td>\n",
       "        <td class=\"number\">11.6%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:49%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">2</td>\n",
       "        <td class=\"number\">46</td>\n",
       "        <td class=\"number\">8.6%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:36%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "        </div>\n",
       "        <div role=\"tabpanel\" class=\"tab-pane col-md-12\"  id=\"extreme-3389567407320575927\">\n",
       "            <p class=\"h4\">Minimum 5 values</p>\n",
       "            \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">0</td>\n",
       "        <td class=\"number\">127</td>\n",
       "        <td class=\"number\">23.7%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">1</td>\n",
       "        <td class=\"number\">81</td>\n",
       "        <td class=\"number\">15.1%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:64%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">2</td>\n",
       "        <td class=\"number\">46</td>\n",
       "        <td class=\"number\">8.6%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:36%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">3</td>\n",
       "        <td class=\"number\">69</td>\n",
       "        <td class=\"number\">12.9%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:54%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">4</td>\n",
       "        <td class=\"number\">71</td>\n",
       "        <td class=\"number\">13.3%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:56%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "            <p class=\"h4\">Maximum 5 values</p>\n",
       "            \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">2</td>\n",
       "        <td class=\"number\">46</td>\n",
       "        <td class=\"number\">8.6%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:58%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">3</td>\n",
       "        <td class=\"number\">69</td>\n",
       "        <td class=\"number\">12.9%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:87%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">4</td>\n",
       "        <td class=\"number\">71</td>\n",
       "        <td class=\"number\">13.3%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:89%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">5</td>\n",
       "        <td class=\"number\">62</td>\n",
       "        <td class=\"number\">11.6%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:78%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">6</td>\n",
       "        <td class=\"number\">79</td>\n",
       "        <td class=\"number\">14.8%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "        </div>\n",
       "    </div>\n",
       "</div>\n",
       "</div>\n",
       "    <div class=\"row headerrow highlight\">\n",
       "        <h1>Correlations</h1>\n",
       "    </div>\n",
       "    <div class=\"row variablerow\">\n",
       "    <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAm0AAAIiCAYAAACaBf8HAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD%2BnaQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3X98zfX///H7zsawXza2JS1EmG28KRGS%2BdGkH5OSH9%2BK9BvDIlEppLzfNfTxo6L3h/jUu/f6pNDyfsdU%2BqX3J%2B/ELAuz8iPZtB2zYT/P9w85b4fJeJ2Xs9fZ7Xq57MKerx/ncV4vcx7uz9frxcfhcDgEAACAGs3m6QIAAABwfjRtAAAAFkDTBgAAYAE0bQAAABZA0wYAAGABNG0AAAAWQNMGAABgATRtAAAAFkDTBgAAYAE0bQAAABbg5%2BkCAFxabdq0OeeyOnXqqGHDhoqNjdXAgQOVkJAgHx%2BfS1gdAOBcfPi/R4Ha5VTT1r59e4WHh7ssKykp0Z49e/TLL79IkuLj4zV//nzVrVv3ktcJAHBF0wbUMqeatkWLFqlv375VrpOenq4nnnhCx44d06hRo/Tkk09eyhIBAFXgmjYAZ%2Bnbt68mTpwoSXrnnXdUUlLi4YoAAFzTBqBKN910k55//nkdP35cP/zwgzp27OiyfNOmTXr77bf1/fffy263KzAwUG3bttWdd96p2267rcp92u12LVu2TBs3btTPP/%2Bs0tJShYSEqH379rr//vvVpUuXs7bp3bu3Dhw4oDfeeEMlJSWaN2%2Be9u3bpzfffFPXXHONJGnnzp1aunSpNm/erEOHDsnPz08RERG67rrrdM8991R5Hd/Ro0e1fPlybdiwQT///LPKysrUuHFjderUSSNHjlRcXJzL%2Bvv371efPn0kSdu2bVN2drZee%2B01bdmyRXa7XREREerTp4%2BSk5PVoEGDizrmAPBHSNoAVKlhw4bO3xcWFrosmzdvnkaOHKn169crMDBQXbp0UVhYmDZt2qRJkyZpwoQJqqiocNkmLy9PgwYN0uuvv649e/aoXbt26tq1q%2BrWratPP/1UI0aM0MqVK89Zz549e5ScnKw6deqoa9euqlevniTp//7v/3TnnXfqgw8%2BUElJia655hp17NhRRUVFevfdd3X33Xdr06ZNLvs6cOCA7rjjDi1YsEC7d%2B9W27Zt1bVrVzkcDqWlpenuu%2B/We%2B%2B9d85avvvuOw0fPlyZmZlq27atWrRooQMHDmjFihVKSkqq9jEGgAviAFCrtG7d2tG6dWvH%2BvXr/3C97Oxs57rbtm1zjq9bt87RunVrR%2BfOnR3/%2Bte/XLb5v//7P0e3bt0crVu3dixbtsxl2axZsxytW7d2JCQkOA4fPuwcLy8vd7zwwguO1q1bO6655hpHUVGRy3bx8fGO1q1bO3r37u147bXXzqpz8ODBjtatWzv%2B/Oc/OyoqKlz2O2/ePEfr1q0dt956q8s2w4cPd7Ru3doxaNAgx6FDh5zjFRUVjgULFjhat27tiI2NdezZs8e5bN%2B%2Bfc7j0atXL8eSJUsclZWVzuUffPCBc/mOHTv%2B6NACwEUhaQNQpXXr1kmSQkJCFB0d7Rx/7bXXJElPPvmkrrvuOpdtOnfurMmTJ0uSli9f7rIsIiJCt9xyi8aOHatGjRo5x319fZWcnCybzaajR4/q%2B%2B%2B/r7KeyspKPfzww2eN79ixQ5I0aNAg2Wz/%2BSvN19dX48aN07hx43TvvfeqtLRUkrR161Zt3rxZkvTSSy8pIiLCuY3NZtPYsWMVExOj0tJS/f3vf6%2BylquvvloPPfSQy%2BNQEhMTFRYW5nwNAHA3rmkDcJaNGzdq8eLFkqQHH3xQfn4n/6o4dOiQMjMzJUn9%2BvWrctu%2BffvKx8dHv/zyi3JyctSiRQtJ0kMPPXTO16tfv74aNWqkvLw85eXlVblO165dXZqyU4KDg3X48GGtXbtW48ePd1lms9k0ZswYl7EvvvhC0sm7aFu2bHnO95CZmXnWtOopVV2z5%2BPjo6ioKOXn56ugoKDK7QDACJo2oJZavHix3n//fZex0tJS/fTTT9q3b5%2Bkk%2BnVgw8%2B6Fy%2Bc%2BdO5%2B%2BnTJlyzn37%2BfmprKxMP/30k7NpO7X/jRs3avv27crNzVVhYaEcvz916OjRo5JOJmpViYyMrHJ8xIgRmjNnjl599VV98cUXuvXWW3X99def8yHCu3fvlnQyLTuXq666StLJ6%2BiqcuWVV1Y57u/vL0kqKys7574B4GLRtAG11LZt284a8/X1VWhoqPr06aMhQ4boxhtvdFl%2B5MgR5%2B83bNhw3tc41YhJUlZWlsaMGaP9%2B/dfVL0hISFVjp%2BaMl28eLEyMjKUkZEhSQoPD1f//v01YsQIRUVFnVVTcHDwOV8rKChI0snm68SJE86bHk7hYcMAPIGmDail/ujhuudy6hquOnXqKCMjo9r/xdWJEyf02GOP6ZdfflHz5s01evRoXX/99QoNDVWdOnUk/efRHud77ao8/PDDGjp0qNLT07Vx40Z9/fXXysvL0//8z/8oNTVVL730km6%2B%2BeZqv0/Hac8cr2pKFgA8gaYNQLWdegxIWVmZCgoKnBfen8/nn3%2BuX375RT4%2BPlqyZImaNWt21jonTpwwVFtwcLAGDRqkQYMGqaKiQl9//bUWLVqkLVu26Omnn1bXrl0VGhrqTOzOfIzJ6U4tq1evHqkagBqDf0ICqLbWrVs7f79r165qb/fTTz9JOnktWFUN2969e/Xbb78Zru8UX19f3XDDDVqxYoUiIyNVXFzsnA4%2BdS3b6dfnnenUsj%2B67g0ALjWaNgDVFh4erpiYGEnSu%2B%2B%2BW%2BU6OTk5SkxMdD4aRPrPNWLn%2Bu%2BwFi1a5Pz9mQ/l/SNbt27VU089paVLl1a5vG7dumrcuLGk/yR5p67T27lzZ5WNW2VlpfNxJzfccEO1awEAs9G0Abggjz76qCQpLS1Nb775psv1Xz///LPGjRunrKwsHTx40Dnetm1bSdKvv/6q9PR05/jx48c1a9YsZWRkOP%2BbrHPdsVkVX19frVy5UvPnz9dnn3121vL169drx44dqlOnjjp16iRJateunXr06CFJeuqpp5Sfn%2B9cv7y8XC%2B//LKys7MVFBSkoUOHVrsWADAb17QBuCA33XSTHn74YS1ZskSzZ8/WihUrdNVVV6mgoEA7duxQRUWF2rVrp0mTJjm36dixo3r06KEvv/xSSUlJ6tChg/z9/bV9%2B3b5%2Bflp2bJl%2Buijj7RlyxatWLFCP/74ox577DF17tz5D2uJjY3VqFGjtHTpUj3yyCNq2rSpmjVrJpvNpv379zunZZ9%2B%2BmmFh4c7t3vxxRc1YsQIZWRkqHfv3mrdurX8/f21a9cuFRQUqF69epozZ845HzMCAJ5A0wbggk2cOFHdunXTW2%2B9pe%2B//16bNm2Sv7%2B/YmNjNWDAAA0fPvysC/hfeeUVpaSkaMOGDdq%2BfbsiIiKUkJCgRx99VFdeeaUuu%2Bwy/fDDD/ruu%2B%2B0a9euat%2BZ%2BuSTT6pz585atWqVtm/fri1btqi8vFzh4eG69dZbdc8995z1n91HRkbqvffe05tvvqn169dr165dKi8vV2RkpG666SY98MADVV57BwCe5OM4fW4DAAAANRLXtAEAAFgATRsAAIAF0LQBAABYAE0bAACotb744gt169ZNycnJf7heZWWl5s2bpz59%2Bqhz58564IEHtG/fPudyu92uCRMmqFu3burRo4eefvppw//Ty5lo2gAAQK30xhtvaNasWdW6W/ztt9/Whx9%2BqCVLlujTTz9V8%2BbNNWbMGOezKqdNm6bjx48rLS1NK1euVHZ2tlJSUtxaL00bAAColfz9/fXee%2B9Vq2lLTU3VyJEj1bJlSwUGBio5OVnZ2dnaunWrDh8%2BrPT0dCUnJyssLEyRkZEaPXq0Vq5cqbKyMrfVy3PaAACAZeXm5iovL89lLDw8XBEREefd9r777qvWa5w4cUK7d%2B9Wu3btnGOBgYFq1qyZMjIydPToUfn6%2BqpNmzbO5TExMTp27Jj27NnjMm4ETZtZqvlgUMto0ULatUu6%2BmopJ8fT1biP3e7pCtzLx0cKCpKOHpW86BGMeaXeNSlgs/koLCxA%2BfnFqqz0nvNUt26Qp0twGy/9UZIkhYR46IVN%2BlxMnT9fCxcudBkbO3askpKS3PYaR44ckcPhUMgZBy8kJEQFBQVq2LChAgMDXR4KfmrdgoICt9VB04bqadhQ8vU9%2BStqLh%2Bf/3x52yeNF7HZfOTj4yObzcermjZvwo%2BSdQwZMkS9e/d2GTv9v61zpz/6/wguxf9VQNMGAADMZzMnMY%2BIiKjWVKgRDRs2lM1mk/2M2Rm73a5GjRopLCxMRUVFqqiokK%2Bvr3OZJDVq1MhtdXjXnAMAAKiZbDZzvi4Bf39/XX311crMzHSOFRYWau/evWrfvr2io6PlcDiUlZXlXJ6RkaHg4GC1aNHCbXXQtAEAAJzh0KFD6t%2B/v/NZbMOGDdOKFSuUnZ2toqIipaSkKDo6WnFxcQoLC1NCQoJeeeUV5efn69dff9WiRYt01113yc/PfZOaTI8CAADzXaJU7ELExcVJksrLyyVJ6enpkk6mZGVlZcrJyVFpaakkaejQocrLy9O9996r4uJidenSxeUGiJkzZ%2Bq5555Tnz59VKdOHd16663nfWDvhfJxXIor52ojb7t7tGNH6bvvpE6dpC1bPF2N%2B3jb3aM2239ueaus9HQ1buNtd4/6%2BdkUGhqggoJilZd7z3nyprtHvfRHSZIH7x719zdnvyUl5uy3BiJpAwAA5quBSZvV0LQBAADz0bQZxhEEAACwAJI2AABgPpI2wziCAAAAFkDSBgAAzEfSZhhNGwAAMB9Nm2EcQQAAAAsgaQMAAOYjaTOMIwgAAGABJG0AAMB8JG2G0bQBAADz0bQZxhEEAACwAJI2AABgPpI2wziCAAAAFkDSBgAAzEfSZhhNGwAAMB9Nm2EcQQAAAAsgaQMAAOYjaTOMIwgAAGABJG0AAMB8JG2G0bQBAADz0bQZxhEEAACwAJI2AABgPpI2wziCAAAAFkDSBgAAzEfSZhhNGwAAMB9Nm2EcQQAAAAsgaQMAAOYjaTOMIwgAAGABJG0AAMB8JG2G0bQBAADz0bQZxhEEAACwAJI2AABgPpI2wziCAAAAFkDSBgAAzEfSZhhNGwAAMB9Nm2EcQQAAAAsgaQMAAOYjaTOMpg0AAJiPps0wjiAAAIAFkLQBAADzkbQZRtMGAABqrQMHDmjGjBnaunWrGjRooAEDBmjixImyndFkjho1St9%2B%2B63LWHl5ucaMGaOxY8fq3nvv1XfffeeyXYsWLbRmzRq31UrTBgAAzFdDk7akpCTFxMQoPT1dv/32mx555BE1btxY999/v8t6S5cudfm%2BsLBQAwYMUL9%2B/Zxjzz//vAYNGmRarTXzCF5CI0aM0J///GeXsUWLFmno0KEeqggAAC9ks5nzZUBGRoaysrI0adIkBQUFqXnz5ho5cqRSU1PPu%2B0rr7yifv36qU2bNoZquBC1PmkbOHCg5s6dq8mTJzsjzXXr1unuu%2B%2Bu9j5yc3OVl5fnMhbeooUiGjZ0a60e1bat66/eoob%2By%2B%2BinXo/Xva%2B/Py86/34%2BtpcfvUW3vTHzkt/lFRZ6ekK3K/Kz%2BDwcEVERJx328zMTDVt2lQhISHOsZiYGOXk5KioqEiBgYFVbvfzzz9r1apVSk9Pdxlfu3at/vrXv%2BrgwYPq0KGDZs6cqSuvvPIi3lXVan3TdtNNN2nmzJn617/%2Bpeuvv1779u1Tdna2br755mrvIzU1VQsXLnQZGzt%2BvJLGj3d3uZ73t795ugJUR0CApytwq1BPF2CS4OD6ni4B5%2BFlP0o6csSDL25SB1zlZ/DYsUpKSjrvtna7XcHBwS5jpxq4goKCczZtS5Ys0Z133qmwsDDnWMuWLVW/fn2lpKSosrJSs2bN0oMPPqi0tDTVrVv3Qt9WlWp90xYQEKC%2BfftqzZo1uv7667Vu3Tp1797d5UScz5AhQ9S7d2%2BXsfDbbpOWL3d3uZ7Ttu3Jhm34cCkry9PVuM/GjZ6uwL1stpOfMsXFXvVP6oJy74o7fH1tCg6ur8LC46qo8J7z5OfnPR2Ol/4oeaUqP4PDw6u9vcPhuKDXs9vtWr16tf7xj3%2B4jE%2BfPt3l%2B5kzZ6pLly7697//reuvv/6CXuNcan3TJp2cIh03bpxmzJih9evX65577rmg7SMiIs6OYXNy3FhhDZKVJW3Z4ukq3Mdb/zaurPSq91Ze7ukKzFFRUanycu85T942lSh53Y%2BSZ5n0B6TKz%2BBqCgsLk91udxmz2%2B3y8fE5Z3izYcMGtWjRQlFRUX%2B478DAQIWEhOjQoUMXVVtVvPBH7MJdf/31CggI0P/%2B7/9q165d6tOnj6dLAgDAu9TAGxFiY2N18OBB5efnO8cyMjLUqlUrBZxjbnzDhg3q3r27y1hRUZGmT5/u0qDl5%2BcrPz//vM3dhaBpk2Sz2XTbbbdp7ty56tOnj%2BrX5zoTAAC8Xbt27RQXF6c5c%2BaoqKhI2dnZWrZsmYYNGyZJ6t%2B/vzZv3uyyzY4dO3TFFVe4jAUGBmrr1q2aNWuW7Ha7jhw5ohkzZqhNmzbq2LGj2%2BqlafvdwIEDVVRUpNtuu83TpQAA4H1qYNImSfPnz1dubq66d%2B%2Bu%2B%2B67TwMHDtTw4cMlSTk5OTp27JjL%2Bnl5eWrcuPFZ%2B1m0aJEcDocSEhLUq1cvlZWVacmSJWc9pNcIrmn73W%2B//aamTZueFXkCAADvddlll%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%2BPl4vv/yyKisrz1pvwYIFio6OVlxcnMvX4cOHJUklJSV69tln1bNnT3Xp0kXjxo1TQUGBW2ulaQMAAOaz2cz5MigpKUmRkZFKT0/XsmXLlJ6eruXLl1e5bmJiojIyMly%2BGjduLEmaN2%2BeMjMzlZqaqo8//lgOh0NTp041XN/paNoAAID5amDTlpGRoaysLE2aNElBQUFq3ry5Ro4cqdTU1AvaT3l5ud577z2NHj1aTZo0UcOGDTVhwgR99tlnOnTokKEaT8c1bWax2z1dgXud%2BsHYuFGqIja2rIYNPV2Be3XsKH33nXTjjdKWLZ6uxn1yCz1dAaohpF6Jp0twHx8fSXUV5F8qORyersbN/D1dgFvl5uYqLy/PZSw8PFwRERHn3TYzM1NNmzZVSEiIcywmJkY5OTkqKipSYGCgy/o//vijhg4dqp07d6pJkyaaOnWqevToob179%2Bro0aOKiYlxrtuyZUvVq1dPmZmZioyMNPguT6JpAwAA5jPpRoTU1FQtXLjQZWzs2LFKSko677Z2u13BwcEuY6cauIKCApem7bLLLlNUVJQmTpyoiIgIpaam6tFHH9WaNWtk/z2oOXNfwcHBbr2ujaYNAABY1pAhQ9S7d2%2BXsfDw8Gpv76hmkjp48GANHjzY%2Bf3IkSP10Ucfac2aNerZs%2BcF7eti0bQBAADzmZS0RUREVGsqtCphYWHOlOwUu90uHx8fhYWFnXf7pk2bKjc317mu3W5XQECAc/mRI0fUqFGji6qtKtyIAAAAzFcDb0SIjY3VwYMHlZ%2Bf7xzLyMhQq1atXJovSXr11Ve1adMml7Hs7GxFRUUpKipKISEhyszMdC7buXOnSktLFRsba6jG09G0AQCAWqldu3aKi4vTnDlzVFRUpOzsbC1btkzDhg2TJPXv31%2BbN2%2BWdDJFmzFjhvbs2aOSkhItXbpUe/fu1R133CFfX1/dfffdev3113Xw4EEVFBRo7ty56tevn/ORIO7A9CgAADBfDf0fEebPn69p06ape/fuCgwM1NChQzV8%2BHBJUk5Ojo4dOyZJmjhxoqST17LZ7Xa1atVKb775pi677DJJ0rhx41RcXKzExESVl5crPj5e06dPd2utPg6zr5qrrY4c8XQF7mWzSUFB0tGjPPKjJjv1yI9OnbzqkR95XvbIDz8/m0JDA1RQUKzycu/5eQoPruvpEtzHx0eqW1cq9cJHfvh76JEfs2ebs183P8C2JiNpAwAA5quhSZuV0LQBAADz0bQZxhEEAACwAJI2AABgPpI2wziCAAAAFkDSBgAAzEfSZhhNGwAAMB9Nm2EcQQAAAAsgaQMAAOYjaTOMIwgAAGABJG0AAMB8JG2G0bQBAADz0bQZxhEEAACwAJI2AABgPpI2wziCAAAAFkDSBgAAzEfSZhhNGwAAMB9Nm2EcQQAAAAsgaQMAAOYjaTOMIwgAAGABJG0AAMB8JG2G0bQBAADz0bQZxhEEAACwAJI2AABgPpI2w2jaAACA%2BWjaDOMIAgAAWABJGwAAMB9Jm2EcQQAAAAsgaQMAAOYjaTOMpg0AAJiPps0wjiAAAIAFkLQBAADzkbQZxhEEAACwAJI2AABgPpI2w2jaAACA%2BWjaDOMIAgAAWABJGwAAMB9Jm2EcQQAAAAsgaQMAAOYjaTOMpg0AAJiPps0wjiAAAIAF1Jimbf/%2B/WrTpo2ys7Mv6evee%2B%2B9SklJuaSvCQBArWOzmfNVi9Sudytp3759%2Buc//%2BnpMgAAQA1w4MABPfzww%2BrSpYvi4%2BP18ssvq7Kyssp133nnHSUkJKhjx45KTExUenq6c9mUKVPUrl07xcXFOb%2BuvfZat9Za665pW7dunbZv367%2B/ft7uhQAAGqPGpqKJSUlKSYmRunp6frtt9/0yCOPqHHjxrr//vtd1vv44481Z84cLV68WO3bt9eqVas0YcIE/eMf/1BUVJQk6bHHHlNSUpJptdbIps1ut2vWrFn65ptvVFxcrK5du2r69OmKjIzU/v371adPHy1dulQvv/yyfvrpJ7Vt21YpKSm64oorJEmvvvqqli1bJj8/Pz3yyCPauHGjOnXqpAYNGjinQtPT0/X9999LkioqKvTss88qLS1N/v7%2BmjZtmgYMGFDtenNzc5WXl%2BcyFl6/viLCw910RGqAUz9sNfSH7qJ17OjpCtyrbVvXX72En593/bnz9bW5/Oo1fHw8XYH7nHov3vSeJMnh8Nxrm/T5UeVncHi4IiIizrttRkaGsrKytGzZMgUFBSkoKEgjR47U8uXLz2raTpw4occff1zXXHONJGnw4MFKSUnR999/72zazFYjm7YpU6bIz89PH330kXx9ffXcc89p6tSpWrp0qXOdFStWaPHixfL399d9992nv/71r5o%2BfbrWr1%2Bv119/XUuXLlW7du00e/ZsZWZmqlOnTnrggQe0a9culZSUaN68ec59paWl6cUXX9QzzzyjhQsXavr06brpppvk51e9w5OamqqFCxe6jI0dM0ZJ48a554DUJAEBnq7Avb77ztMVmONvf/N0BW4V6ukCTBIcXN/TJeB86tTxdAXuVVLi6QrcrsrP4LFjq5V4ZWZmqmnTpgoJCXGOxcTEKCcnR0VFRQoMDHSOJyYmumxbWFio4uJiRUZGOse%2B%2BeYbbdiwQT///LNatmyp6dOnKzY29mLf2llqXNOWn5%2BvTz/9VGvXrnUexEmTJqlXr14unfSwYcOcB6pHjx7KyMiQJG3cuFE9evRwziNPnjxZq1ev/sPX7NSpk2644QZJUv/%2B/bV48WLl5%2BdXq0uXpCFDhqh3794uY%2BH160tHj1Zre0uw2U42bMXF0jnm%2Bi3pxhs9XYF7tW17smEbPlzKyvJ0NW5TsOELT5fgVr6%2BNgUH11dh4XFVVHjPz1NogBc1OD4%2BJxu2sjLPplPexKSkrcrP4GrOdNntdgUHB7uMneo9CgoKXJq20zkcDj3zzDPq0KGDrrvuOklSVFSUbDabxo8fr4CAAC1cuFCjRo3Sxx9/rNBQ9/zTs8Y1bXV%2B/1fNwIEDXcZ9fX118OBBhYWFSZJzKlSS6tevr5Lf//WQl5enK6%2B80rksKChIzZs3/8PXPH1f/v7%2BkqTS0tJq1xwREXF2g3fkiHc1N6dUVnrX%2B9qyxdMVmCMry6veW3m5F/2ZO01FRaV3vTdvbG4cDu98X16kys/gC%2BC4wPNbVlamKVOmaPfu3VqxYoVzfMyYMS7rPfHEE0pLS1N6eroGDx580fWdrsY1bfXq1ZMkff7551V2pvv375ck%2BZzjOoPKysqzpjVt5%2Bnuz7UvAADgJjXwmuiwsDDZ7XaXMbvdLh8fH2dIdLoTJ05o9OjROn78uN5%2B%2B%2B0/TNB8fX3VpEkT5ebmuq3emncEdbLJ%2BvHHH53fl5WV6dChQ9XatlGjRvrll1%2Bc3xcVFSknJ8ftNQIAgAtQA5/TFhsbq4MHDyo/P985lpGRoVatWingjGu4HQ6HkpOT5efnpzfffNOlYXM4HJo9e7ayTrsspbS0VHv37nXrTQo1rmmrU6eOBgwYoJSUFP366686ceKE5s6dq1GjRlUrwuzatas%2B//xzbdu2TSdOnNBLL73kTO%2Bkk9OfBw8eVGFhocrLy818KwAAoAY79Vy1OXPmqKioSNnZ2Vq2bJmGDRsm6eR17ps3b5Ykffjhh9q9e7f%2B67/%2By3kp1Sk%2BPj7av3%2B/ZsyYoUOHDqm4uFgpKSmqU6eO%2Bvbt67Z6a1zTJknTpk1Ts2ZC9a8iAAAgAElEQVTNdMstt%2BiGG27Q7t279eqrr1ZrGvP222/XnXfeqfvuu08JCQnq0KGDrrzySue2t912m3JychQfH%2B/WyBIAAPyBGpi0SdL8%2BfOVm5ur7t2767777tPAgQM1fPhwSVJOTo6OHTsmSVq5cqUOHDig6667zuUBus8884wk6YUXXlDz5s01aNAgdevWTTt27NDy5cvVoEEDwzWe4uO40CvwLKC0tFR169Z1fh8fH6/Ro0e77ULAajly5NK91qVgs0lBQSfviPWmGxEaNvR0Be7VsePJx5h06uRVNyLk5RZ6ugS38vOzKTQ0QAUFxV51I0J4cN3zr2QVPj5S3bpSaan33YhwRkp0yWzaZM5%2Br7/enP3WQDUyaTPi22%2B/VefOnbVt2zZVVFTo/fffV15enq6vRScVAIAap4YmbVZS4%2B4eNapz585KTk7WhAkTlJ%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%2BPl4vv/yyKisrq1x3xYoVSkhIUKdOnTRs2DBt377duaykpETPPvusevbsqS5dumjcuHEqKChwa600bQAAwHw2mzlfBiUlJSkyMlLp6elatmyZ0tPTtXz58rPW%2B%2BSTT7RgwQK99NJL%2BvrrrxUfH69HH31Ux44dkyTNmzdPmZmZSk1N1ccffyyHw6GpU6caru90NG0AAMB8JjVtubm5yszMdPnKzc2tVkkZGRnKysrSpEmTFBQUpObNm2vkyJFKTU09a93U1FQNGjRIHTp0UL169fTggw9Kkj799FOVl5frvffe0%2BjRo9WkSRM1bNhQEyZM0GeffaZDhw657RByTZtJ8kq9qx/287MpVFJBuU3l5Z6uxo1yCz1dgVs5z9OGL1ReXnW8D5glr7DU0yW4jZ%2BfTaF166qguMyrfpbCw4M8XYLbpaamauHChS5jY8eOVVJS0nm3zczMVNOmTRUSEuIci4mJUU5OjoqKihQYGOiy7oABA5zf22w2RUdHKyMjQ9HR0Tp69KhiYmKcy1u2bKl69eopMzNTkZGRRt6iE00bAAAwnUM%2Bpux3yJAh6t27t8tYeHh4tba12%2B0KDg52GTvVwBUUFLg0bXa73aW5O7VuQUGB7Ha7JJ21r%2BDgYLde10bTBgAALCsiIkIREREXvb3D4XDbuheyr4vhXXN4AACgRqqsNOfLiLCwMGdKdordbpePj4/CwsJcxkNDQ6tcNywszLnumcuPHDmiRo0aGSvyNDRtAADAdDWxaYuNjdXBgweVn5/vHMvIyFCrVq0UEBBw1rqZmZnO7ysqKvTDDz%2BoQ4cOioqKUkhIiMvynTt3qrS0VLGxscaKPA1NGwAAqJXatWunuLg4zZkzR0VFRcrOztayZcs0bNgwSVL//v21efNmSdKwYcO0atUqff/99zp%2B/Lhee%2B011a1bV7169ZKvr6/uvvtuvf766zp48KAKCgo0d%2B5c9evXT40bN3ZbvVzTBgAATGc0FTPL/PnzNW3aNHXv3l2BgYEaOnSohg8fLknKyclxPoetZ8%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%2B1Sa1t2pYvX64//elPSktL83QpAACgBrLb7ZowYYK6deumHj166Omnn9aJEyfOuf66det0%2B%2B23q2PHjkpISNC7777rXLZgwQJFR0crLi7O5evw4cPVrqfWXtNWVFSkDh06yGartX0rAACXjBVTsWnTpqm0tFRpaWkqKyvT%2BPHjlZKSomeeeeasdbdt26ZJkyZp7ty56tWrl7766iuNGTNGV111la699lpJUmJiov785z9fdD21smm75557tHnzZv373//WunXr1Lt3b6WlpamoqEhXXnmlnnjiCd1www3V3l9ubq7y8vJcxvz8GigiIsLdpXuMr6/N5VfUTJwna%2BA81XycI/czq2mr6jM4PDzc8Gfw4cOHlZ6erg8%2B%2BEBhYWGSpNGjR2v8%2BPF68sknVadOHZf17Xa7HnnkEfXt21eSdOONN6p169bavHmzs2kzqlY2bW%2B99ZbuvfdedejQQW3atNEbb7yhNWvWKCQkRKtWrdKTTz6pjRs3nnVCziU1NVULFy50GRszZozGjRtnRvkeFRxc39MloBo4T9bAear5OEc1X1WfwWPHjlVSUpKh/e7YsUO%2Bvr5q06aNcywmJkbHjh3Tnj17XMYlqWfPnurZs6fz%2B/LycuXl5SkyMtI59uOPP2ro0KHauXOnmjRpoqlTp6pHjx7VrqlWNm2nKywslJ%2Bfn%2BrXry9fX1/deeeduuOOOy5o2nTIkCHq3bu3y5ifXwMVFBS7u1yP8fW1KTi4vgoLj6uiwoIZdy3BebIGzlPN563nKDQ0wGOvbVbSVtVncHh4uOH92u12BQYGysfHxzkWEhIiSSooKDjv9ikpKWrQoIHzhsfLLrtMUVFRmjhxoiIiIpSamqpHH31Ua9as0VVXXVWtmmp903bLLbdo9erV6tmzp7p3765evXrplltuuaCmLSIi4qwYNi/vqMrLvecH/ZSKikqvfF/ehvNkDZynmo9zVPNV9RlcXatXr9bkyZOrXJacnCyHw3HB%2B3Q4HEpJSVFaWppWrFghf39/SdLgwYM1ePBg53ojR47URx99pDVr1mjChAnV2netb9oaNmyod999V999950%2B/fRTzZ8/X%2B%2B8847efvtt%2BfnV%2BsMDAIBb1MQbERITE5WYmFjlsq%2B%2B%2BkpFRUWqqKiQr6%2BvpJPpmyQ1atSoym0qKys1depUbdu2Te%2B8846ioqL%2B8PWbNm2q3Nzcatdb66%2BwLCkp0fHjx9WpUydNnDhRaWlp2rlzp7KysjxdGgAAXsNqz2mLjo6Ww%2BFw6QcyMjIUHBysFi1aVLnNiy%2B%2BqF27dlXZsL366qvatGmTy1h2dvZ5G7vT1fqm7YUXXtCTTz6p/Px8ORwOZWZmqrKyUpdffrmnSwMAAB4SFhamhIQEvfLKK8rPz9evv/6qRYsW6a677nLOxI0YMUJr166VJP373//WmjVrtGTJEjVs2PCs/dntds2YMUN79uxRSUmJli5dqr179%2BqOO%2B6odk21fv5v4sSJeu6555SQkKDy8nI1a9ZMc%2BbMcd7eCwAAjKuJ06PnM3PmTD333HPq06eP6tSpo1tvvVXJycnO5fv27dORI0ckSStXrtTRo0cVHx/vso/OnTtr6dKlmjhxoqST17LZ7Xa1atVKb775pi677LJq1%2BPjuJir7HBeeXlHPV2CW/n52RQaGqCCgmIuyq3BOE/WwHmq%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%2BffsuqB6aNgAAYDorNm3Tpk3T8ePHlZaWppUrVyo7O1spKSnnXL9z587KyMhw%2BWrfvr0k6e2339aHH36oJUuW6NNPP1Xz5s01ZswYORyOatdD0wYAAHCGw4cPKz09XcnJyQoLC1NkZKRGjx6tlStXqqys7IL3l5qaqpEjR6ply5YKDAxUcnKysrOztXXr1mrvgxsRAACA6cxKxXJzc5WXl%2BcyFh4eroiICEP73bFjh3x9fdWmTRvnWExMjI4dO6Y9e/a4jJ9y8OBB3X///dq%2BfbuCg4M1btw4JSYm6sSJE9q9e7fatWvnXDcwMFDNmjVTRkaG/vSnP1WrJpo2k9StG%2BTpEtzK9nsm6%2BcX4Py9NwipV%2BLpEtzLx0eSFBpQR7qAyL2myyss9XQJqIbwiGBPl%2BA%2BHTtK332n0D43SFu2eLoa9/KivxukkwnWwoULXcbGjh2rpKQkQ/u12%2B0KDAyUz%2B9/r0pSSEiIJKmgoOCs9cPCwtS8eXM9/vjjatWqldavX6/JkycrIiJCV111lRwOh3P70/dX1b7OhaYNAACYzqykbciQIerdu7fLWHh4eLW2Xb16tSZPnlzlsuTk5Au63qxXr17q1auX8/tbbrlF69ev1/vvv69JkyZJ0gXtryo0bQAAwHRmNW0REREXPRWamJioxMTEKpd99dVXKioqUkVFhXx9fSWdTN8kqVGjRtXaf9OmTbV9%2B3Y1bNhQNpvNuf0pdru92vuSuBEBAADgLNHR0XI4HMrKynKOZWRkKDg4WC1atDhr/XfeeUdr1651GcvOzlZUVJT8/f119dVXKzMz07mssLBQe/fudd5dWh00bQAAwHRWe%2BRHWFiYEhIS9Morryg/P1%2B//vqrFi1apLvuukt%2BficnKkeMGOFs1EpLS/X8888rIyNDZWVlSktL0%2Beff66hQ4dKkoYNG6YVK1YoOztbRUVFSklJUXR0tOLi4qpdE9OjAAAAVZg5c6aee%2B459enTR3Xq1NGtt96q5ORk5/J9%2B/bpyJEjkqT77rtPxcXFGj9%2BvPLy8nTFFVdo0aJFio2NlSQNHTpUeXl5uvfee1VcXKwuXbqcdQPF%2Bfg4jF4Vhyr9fg69hs0mBQVJR4961/8f55V3j9atK5WWetUdYt5296ifn02hoQEqKChWebn3/EB5492j6tSJu0fdZOpUc/Y7e7Y5%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%2BXXXnvtBdVzQdOjTZs2VUZGxgW9AAAAgBWnR6dNm6bS0lKlpaWprKxM48ePV0pKip555pmz1p01a5ZmzZrl/L68vFwDBw5U//79nWOPPfaYkpKSLroepkcBAADOcPjwYaWnpys5OVlhYWGKjIzU6NGjtXLlSpWVlZ13%2B%2BXLl%2Bvyyy/XjTfe6LaaLihp279/v/r06aO1a9fqoYce0mOPPab169fr22%2B/VaNGjTR9%2BnT16NFDkvTJJ5/oL3/5i3Jzc3XjjTeqcePG1X6dKVOmKCAgQH5%2Bfvrggw9ks9n0wAMP6KGHHpIktWnTRm%2B88YZ69uwpSXrnnXf0xhtv6JNPPnHWuHjxYv3lL3/RL7/8ogEDBuixxx7T5MmTlZWVpbi4OC1cuFAhISGSJIfDoVmzZmnVqlUKDAzUAw88cEFTubm5ucrLy3MZq18/XOHhEdXeR01ns7n%2B6jV8fDxdgXudej9e9r78/LzrD56vr83lV6/RsaOnK3Cftm1df/UWW7Z47KXNStqq%2BgwODw9XRISxz%2BAdO3bI19dXbdq0cY7FxMTo2LFj2rNnj8v4mQoLC/X666/rb3/7m8v4N998ow0bNujnn39Wy5YtNX36dMXGxla7JkN3j/73f/%2B3XnrpJbVt21bTp0/Xiy%2B%2BqLVr16qwsFDJycmaNGmShgwZok2bNmnixImKjo6u9r7T0tI0ZcoUffXVV1qzZo2mTZumxMTEap%2BEVatW6d1331VWVpbuuece/fzzz3r55ZdVr1493XHHHVq5cqVGjRolSfr888/1xBNP6Ouvv9Znn32mpKQkdejQQe3bt6/Wa6WmpmrhwoUuY2PGjNW4cRcfgdZUAQGersDd6nq6AHPUqePpCtwqtK53nqfg4PqeLsG9vvvO0xW43xkfupbnZf%2Bgk6r%2BDB47dqyhaUjp5PVsgYGB8jntmJ0KewoKCv5w27feekudO3fW1Vdf7RyLioqSzWbT%2BPHjFRAQoIULF2rUqFH6%2BOOPFRoaWq2aDDVt8fHxzsYmISFBq1atUmVlpb788ks1aNBA/%2B///T/ZbDbdeOONuvbaa1VcXFztfV9xxRW64447JEkDBgzQU089pZ9%2B%2BqnaTdudd96poKAgde7cWUFBQerevbuioqIkSe3bt9dPP/3kXDciIkLDhg2TJN10002Kjo7W559/Xu2mbciQIerdu7fLWP364Tp6tFqbW4LNdrJhKy625nUJ5xLkX%2BrpEtzLx%2Bdkw1ZWJjkcnq7GbQqKzz8VYSW%2BvjYFB9dXYeFxVVR4zw9UaJ8bPF2C%2B7Rte7JhGz5cysrydDVewazPjqo%2Bg8PDw6u17erVqzV58uQqlyUnJ8txEX%2BPVlRU6O2339acOXNcxseMGePy/RNPPKG0tDSlp6dr8ODB1dq3oabtiiuucP6%2BXr16qqioUFlZmX799Vc1adJEttPm0po3b67MzMyL2nf9%2Bif/NXquOzaq0qRJE%2Bfv/f39FRkZ6fJ9ael/PqxbtWrlsu2VV16pQ4cOVfu1IiIizmomjxzxrubmlEtxt84l5UWNjQuHw6veW3m5N/2h%2B4%2BKikrvem8enHozTVaWd74vDzDrs6Oqz%2BDqSkxMVGJiYpXLvvrqKxUVFamiokK%2Bvr6STqZvktSoUaNz7vPbb79VaWnpee8M9fX1VZMmTZSbm1vteg1dUGE7xwVOpaWlqqiocBmrvMCzda59V6WqffucEQH/0f7OXOZwOOTv71/t1wcAAN4lOjpaDodDWaclrRkZGQoODlaLFi3Oud2GDRvUtWtX%2Bfn9JxdzOByaPXu2y75KS0u1d%2B9e5yxgdZhyFWxERIQOHTrkEitmZ2e7bf9169Z1Sd327t1raH9nPndu7969hi9gBAAA/2G157SFhYUpISFBr7zyivLz8/Xrr79q0aJFuuuuu5wN2YgRI7R27VqX7Xbs2OEyWyidDJL279%2BvGTNm6NChQyouLlZKSorq1Kmjvn37VrsmU5q2bt26qaioSH//%2B99VWlqq9PR0bd261W37b968udLT01VeXq6MjAx99tlnhva3f/9%2BffDBByorK9OGDRv0448/ql%2B/fu4pFgAAWNLMmTMVFBSkPn366Pbbb1f79u1dHsi/b98%2BHTlyxGWbvLy8Kp%2BY8cILL6h58%2BYaNGiQunXrph07dmj58uVq0KBBtesx5f8eveyyyzRnzhylpKToL3/5i3r27Knhw4dri5uuC3jqqaf07LPP6tprr9W1116rUaNGafHixRe9v4SEBG3dulWzZs1SUFCQZs6c%2BYfRJwAAuDBWvB46KChIc%2BfOPefyTz755Kyxjz/%2BuMp1GzZsqNmzZxuqx8dxMbdG4LzOaLwtz2aTgoKko0et%2BYN3LiH1Sjxdgnv5%2BEh160qlpV51I0JeoXfd5evnZ1NoaIAKCoq96kaE8IhgT5fgPh07nnyESadO3ncjgof%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%2BzJaRkaF%2B/frp7rvvPu%2B6K1asUEJCgjp16qRhw4Zp%2B/btzmUlJSV69tln1bNnT3Xp0kXjxo1TQUHBBdVC0wYAAExnxaZtzZo1SkpKUrNmzc677ieffKIFCxbopZde0tdff634%2BHg9%2BuijOnbsmCRp3rx5yszMVGpqqj7%2B%2BGM5HA5NnTr1guqhaQMAAKhCSUmJUlNT1aFDh/Oum5qaqkGDBqlDhw6qV6%2BeHnzwQUnSp59%2BqvLycr333nsaPXq0mjRpooYNG2rChAn67LPPdOjQoWrXw40IAADAdGalYrm5ucrLy3MZCw8PV0REhOF9Dx48uNrrZmZmasCAAc7vbTaboqOjlZGRoejoaB09elQxMTHO5S1btlS9evWUmZmpyMjIar0GTZtJQkI8XYF75ebm6s03UzVkyBC3/CDUHP6eLsCtcnNzlZrqfecpPNz7ztOCBUu97jzJ4fB0BW6Tm5ur1AULNOSf//Suc%2BRBZv3xWLAgVQsXLnQZGzt2rJKSksx5wXOw2%2B0KOePDPyQkRAUFBbLb7ZKk4OBgl%2BXBwcEXdF0bTRuqJS8vTwsXLlTv3r35C6wG4zxZA%2Bep5uMcWceQIUPUu3dvl7Hw8PBqbbt69WpNnjy5ymWzZ8/WoEGDLqgWx3k60/MtPx%2BaNgAAYFkREREX3VgnJiYqMTHRLXWEhoY6E7VT7Ha7rr76aoWFhTm/DwgIcC4/cuSIGjVqVO3X4EYEAAAAg2JjY5WZmen8vqKiQj/88IM6dOigqKgohYSEuCzfuXOnSktLFRsbW%2B3XoGkDAAC4CP3799fmzZslScOGDdOqVav0/fff6/jx43rttddUt25d9erVS76%2Bvrr77rv1%2Buuv6%2BDBgyooKNDcuXPVr18/NW7cuNqvx/QoqiU8PFxjx46t9nUC8AzOkzVwnmo%2BzhEkKSEhQb/88osqKipUWVmpuLg4SdI///lPNW3aVDk5Oc7nsPXs2VOPP/64JkyYoN9%2B%2B01xcXFasmSJ6tWrJ0kaN26ciouLlZiYqPLycsXHx2v69OkXVI%2BPw%2BhVcQAAADAd06MAAAAWQNMGAABgATRtAAAAFkDTBgAAYAE0bQAAABZA0wYAAGABNG0AAAAWQNMGAABgATRtAAAAFkDThioNGzZMf//732W32z1dCgAAEP%2BNFc5h0aJFWr9%2BvbKzs9W9e3fdfvvt6tOnj/z9/T1dGqpQWVmp/Px8lZaWnrXs8ssv90BFOFNFRYV8fX0lnTxfWVlZatKkiUJDQz1cGQCroGnDH9q3b5/Wr1%2Bv9evXa/fu3erbt69uu%2B02devWzdOl4Xdr1y314FkAAAqWSURBVK7VjBkzVFhY6DLucDjk4%2BOjHTt2eKgynPLll19qypQp%2BvLLL1VeXq577rlHP/74oyRp7ty5io%2BP93CFON327du1ceNGHTp0SP7%2B/oqMjFS/fv3UrFkzT5eGWo6mDdVSWlqq9957T3PnzlVRUZGaNm2qhx56SEOHDvV0abVe9%2B7dddddd6l///5VJqFXXXWVB6rC6QYOHKj77rtPgwYN0urVq7VgwQJ9%2BOGH2r59u2bPnq3333/f0yXid2%2B99ZZmzZql6OhoRUVFSZL27t2rnTt3aubMmbrrrrs8XCFqMz9PF4Ca7ZtvvtGHH36odevWKSAgQEOHDtXAgQN1%2BPBhzZ49W9nZ/7%2B9%2B42psgrgOP57ruzmlaYNF6hcRTY3WWvKHwMJ6hob6IxCNlPnRn%2BoXLPxQvw3aNXWTFet0OmsFlOjN7WW6UZbLBC0cBPcmIwitgwswZG4lOv0Cru3XoDk9V4UN7mHJ76fV3DO8%2BL3RvfjnOec56zeeOMN0zEnNZ/Pp5KSEkVF8c95ojp37pwKCwslSQ0NDXr66aflcrn02GOPqaury2w4BDlw4ID27t2r3NzcoPG6ujq9%2B%2B67lDYYxUEEhPXee%2B/J4/Fo48aNGhwc1J49e1RfX68tW7ZowYIFWrp0qT777DMdOXLEdNRJLz8/X01NTaZj4A6io6PV398vn8%2BnxsbGke3Qv//%2Bm7I9wXi9Xi1btixk3OPxhLyCAEQa/1sgrPb2dm3atEl5eXmaNm1a2GdiY2P16quvRjgZpKH3oG6aNm2aysrKlJqaKrfbLcuygp4tLS2NdDzc5plnntELL7ygKVOmKDExUcnJyfL5fHr77beVlZVlOh5uUVBQoG%2B//VZr1qwJGq%2Burtazzz5rKBUwhNKGsA4dOjSm5zZs2DC%2BQRBWS0tL0O/z5s1TX1%2Bf%2Bvr6gsZvL3CInPz8fFVXV0uSjh07ppKSEvX39ys/P1%2BS5HA49NBDD2nr1q0mY0LS5s2bR34OBAJ6//33dejQIS1YsECWZamzs1Pd3d3Ky8szmBLgIAJgezdPid7O7/fr4sWLmjVrloFU8Hg8SklJ0bx583TgwAEVFxeP%2BiyroWaVlZWN%2Bdldu3aNYxLgzihtgM0tXrxYZ86cCRn3er3KyclRc3OzgVQ4deqUPv/8c129elXNzc1asmRJ2Ocsy1JVVVWE0wGwI7ZHAZuqqalRTU2NBgcHg7Z3burp6Rm5zBWRl5GRoYyMDElSUVGRvvjiC8OJMBZer1fffPONurq6dOPGjaA5y7K0c%2BdOQ8kAShtgW4888ojOnz%2Bv77//Xk6nM2R%2B4cKFYcscIo/CZh%2Bvv/66Ojs7tWjRIr4AgwmH7VHA5u72vhSAsUtJSVFtba1mzpxpOgoQgpU2wOauXbumffv2hZ1zOByKi4tTZmYm3yAFxmDhwoUh26LARMFKG2BzxcXF%2Bvnnn%2BXz%2BTR//nw5HA51dnbK5XJp7ty5I1eB7Nmzh29cAnfx559/qqysTFlZWYqLi5PDEXwH/apVqwwlA1hpA2zP4/HI7XZr27ZtevDBByVJV69e1QcffKDk5GQVFhbq8OHDqqiooLQBd/Hxxx/r9OnT6ujoCPtOG6UNJrHSBthcVlaW6urqNHXq1KDxGzduaMWKFaqvr1cgEFBaWlrIpbwAgiUnJ6uysnLUK1oAk/j2KGBzg4OD%2BuWXX0LGOzo6dPnyZUlSW1ubYmJiIh0NsJ05c%2BYoKSnJdAwgLLZHAZt78cUX9dJLL%2BmJJ55QfHy8oqKi1NPTo%2BPHj2v16tUaGBjQ888/r02bNpmOCkx427dv11tvvaXVq1eHfactMTHRUDKA7VHgf%2BGHH37QiRMndPHiRQUCAc2cOVMZGRkqKCiQZVk6ffo02z3AGIRbZbMsa%2BRzce3t7QZSAUMobQAADOvu7r7jfHx8fISSAKEobYDN9fT0qLKyUmfPng17v9SXX35pIBUA4H7jnTbA5kpLS3X9%2BnVlZ2fL5XKZjgPYWk5OjizLGnW%2Brq4ugmmAYJQ2wOY6Ojp0/PhxTZ8%2B3XQUwPY2bNgQ9Lvf79cff/yhhoYGvfLKK4ZSAUMobYDNzZ8/XwMDA6ZjAP8L69atCzuen5%2BvvXv36rnnnotwIuA/vNMG2NzJkydVVVWl9evXKz4%2BnisKgHHg9/u1ZMkSLqiGUZQ2wOa4ogC4f3766aeQMZ/Pp9raWrW0tKimpsZAKmAIpQ2wOa4oAO6fcH8EOZ1OJSQkqLy8XJmZmQZSAUMobcD/xKVLl3ThwgU9%2BuijpqMAAMYBBxEAm%2Bvt7VV5ebkaGxsVFRWltrY2/fXXX3r55Ze1f/9%2BzZ0713REwFb6%2BvrU1dUln88XMpednW0gETCE0gbY3DvvvKOYmBjV19dr%2BfLlkqSYmBhlZ2drx44d%2BvTTTw0nBOyjsrJSFRUV8vv9IXO8IwrT2B4FbC4tLU0nTpxQdHS0Fi9erDNnzkgaenna4/Ho1KlThhMC9pGZmamtW7dq5cqVmjp1quk4QBBW2gCbc7lcCve315UrV8KuFgAYXSAQUEFBgaZMmWI6ChDCcfdHAExkS5cuVXl5uTo7OyVJ/f39ampqUklJiZYtW2Y2HGAzhYWFqq6uNh0DCIvtUcDmrly5ou3bt6uhoWFkzOFwaOXKlXrzzTc1Y8YMc%2BEAm9mxY4e%2B%2B%2B47ud1uud3ukMuqP/zwQ0PJAEobYEs3V9VudenSJTmdTjmdTknSAw88IIkvIgD3oqys7I7zu3btilASIBSlDbChpKSkka8eSEOn2m7HFxGA8fPJJ5/otddeMx0DkwylDbChu30F4VZ8EQG4/249qQ1ECqdHARuiiAFmsd4BEzg9CgDAPQr3SgIw3ihtAAAANkBpAwAAsAFKGwAAgA1Q2gAAuEccRIAJlDYAAO7RmjVrTEfAJMQ9bQCASa2oqGjMp0GrqqrGOQ0wOu5pAwBMasnJySM/X79%2BXUePHlVaWpoSExMVCAT022%2B/qbW1VevXrzeYEmClDQCAEaWlpVq1apWefPLJoPHa2lpVV1dr9%2B7dhpIBlDYAAEakpqaqqalJUVHBG1GDg4NKT09XS0uLoWQABxEAABgRGxurr776KmT88OHDevjhhw0kAv7DShsAAMOOHTum0tJSRUdHa/bs2fL7/ert7ZXX69VHH32k3Nxc0xExiVHaAAC4hdfr1Y8//qje3l4NDAwoNjZWjz/%2BuOLi4kxHwyRHaQMAIIyBgYGQMafTaSAJMIQrPwAAGHby5Ent3LlTXV1d8vv9IfPt7e0GUgFDWGkDAGBYbm6u0tPTlZeXJ5fLFTKfnp5uIBUwhNIGAMCwlJQUNTc3h1z5AUwEXPkBAMCw9PR0/frrr6ZjAGGx0gYAwLCvv/5aBw8e1FNPPSW32y2HI3htY%2B3atYaSAZQ2AABG5OTkjDpnWZbq6uoimAYIRmkDAExq586dU0JCgiTp999/l2VZoz6bmJgYqVhACEobAGBSW7RokVpbWyVJSUlJYUvbP//8I8uyuPIDRlHaAACTWk9Pj%2BbMmSNJ6u7uvuOz8fHxkYgEhEVpAwAAsAGu/AAAALABShsAAIANUNoAAABsgNIGAABgA5Q2AAAAG6C0AQAA2AClDQAAwAb%2BBeskGktg/vVQAAAAAElFTkSuQmCC\" class=\"center-img\">\n",
       "    <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAm0AAAIiCAYAAACaBf8HAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD%2BnaQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3X98z/X%2B//H73hvGftnYkEjRmFmi5Gdi0hz9mH6I%2BYb1W2xYJE4UUk41Pz5MPziHOHWcdVJoOUemH06lTkqaZRErRDbsjc3Yr/f3D8f7eGcyXu%2BX917v3a6Xy/sye75%2BvJ%2Bv54u9H%2B7P1%2Bs1H4fD4RAAAACqNZunOwAAAIDzo2gDAACwAIo2AAAAC6BoAwAAsACKNgAAAAugaAMAALAAijYAAAALoGgDAACwAIo2AAAAC6BoAwAAsAA/T3cAgPt89913euedd7R582bt3btXxcXFqlOnjiIiItSuXTv1799fsbGx8vHx8XRXAQAXyIffPQpYn8Ph0HPPPae//vWvkqSQkBBFRkYqMDBQhYWF2rZtmwoLCyVJvXv31rx581S7dm1PdhkAcIEo2gAv8Oabb2r69Ony9/fXtGnTdPvtt8vX19e5/MSJE/r73/%2BuF154QRUVFXrkkUc0btw4D/YYAHChuKYN8ALLly%2BXJA0bNkwDBgxwKdgkyd/fX4mJiXr44YclSW%2B88YZOnDhxyfsJALh4FG2AF9i9e7ckqXXr1r%2B7XmJiov785z9r3bp18vf3lyRNnDhRrVu31gsvvKCioiLNnDlTsbGxiomJUZcuXZSUlKTt27efc58bN25UUlKSevTooXbt2qlLly5KTEzUe%2B%2B9d85t7Ha75syZowEDBqhDhw6Kjo5Wt27dNGLECH355ZeVbhMbG6vWrVtrw4YNWrdunfr376%2BYmBh9/fXXkqT58%2BerdevWmjhxokpKSjR//nzFxcXpmmuu0U033aRnn31Wx48flyTt2LFDo0ePVvfu3RUTE6Pbb79dq1evrvR9i4uLtWjRIt177726/vrr1bZtW3Xu3Fn333%2B/1q1bV%2Bk2Q4cOVevWrbV8%2BXIVFRVpzpw5zr506tRJDz/8sLZu3XrO8QGAylC0AV4gLCxMkvTtt9%2Bed70bb7xRDRs2PGtZUVGREhMTtXz5coWGhqpTp04qLy/XunXrNHDgwEqLjDlz5igxMVHr1q1TYGCgOnfurLCwMG3cuFHjx4/X2LFjVV5e7rJNfn6%2B7rrrLr366qvatWuX2rZtqy5duqh27dr66KOPNHz4cK1YseKcx7Br1y6lpKSoVq1a6tKli7P4PNMTTzyhxYsXq2nTprr66quVl5enN954Q08%2B%2BaS2b9%2BuhIQE5eTkqE2bNgoPD9f27dv1xBNP6OOPP3bZT3FxsYYMGaLU1FRlZ2erZcuW6tq1q0JCQvT5558rKSlJ8%2BfPP2dfS0pKlJiYqKVLl6pJkyaKjo5WeXm5NmzYoKFDhzqLbQCoEgcAy3vyyScdkZGRjtatWzteeOEFx4EDBy54244dOzpuu%2B02l22PHj3quOeeexyRkZGOe%2B65x2W7Dz74wBEZGeno1KmT48svv3RZ9p///MfRrVs3R2RkpGPJkiUuy2bMmOGIjIx0xMXFOQ4ePOhsLysrczz33HOOyMhIx3XXXecoLCx02a53796OyMhIR2xsrOOVV1456zjmzZvniIyMdPTo0cNxxx13OPLz853L1q1b5xyffv36OebPn%2B%2BoqKhwOBwOR3l5uWPUqFGOyMhIx4MPPuiyzyVLljgiIyMdXbp0ceTm5la6LCoqyrF7926XZffdd58jMjLS0atXL8eQIUMchw4dci7Ly8tzdO/e3REZGel4/vnnzzoOADgXkjbAC4wdO1ZNmjSRw%2BHQX/7yF/Xq1Uv33Xef5s6dq88//1zFxcXn3UdhYaGmTp2qiIgIZ1tQUJAmTpwo6dTjRH766SfnsldeeUWS9OSTT%2BqGG25w2VenTp00YcIESdLSpUtdlkVEROjWW29VUlKSGjRo4Gz39fVVSkqKbDabjh07ds7U8PSNFOeSl5enadOmuaSJN998s8LDw%2BVwOOTn56dRo0Y5H3tis9k0aNAgSdL333/vsq%2BAgADdfvvtGjFihFq0aOGybPjw4QoPD1d5ebk2btxYaV/y8/M1e/ZsZxIqSeHh4br99tslSVu2bDnncQDAb/GcNsALNG7cWP/4xz/00ksvKSMjQ%2BXl5frqq6/01Vdf6ZVXXlGtWrXUuXNn3XHHHbr11lvl53f2P/2IiAhdd911Z7V36NBBAQEBKioq0rfffqsWLVrowIEDys7OliT17du30j7dfPPN8vHx0b59%2B5Sbm6srr7xSkpw3Q1Smbt26atCggfLz85Wfn1/pOl26dJHNdu7/b4aHh%2Bvaa689q71p06bKz89X7969z3pOXdOmTSVJR44ccWkfOHCgBg4cWOn7%2BPj46PLLL//dvt5www1q1KjRWe3NmzeXJBUUFJzzOADgtyjaAC8RHh6uF198URMmTNC6deu0ceNGbdq0SYcOHVJpaak%2B/fRTffrpp1q0aJFefvllZ%2BFwWmRkZKX7tdlsatq0qbZv3669e/dKksuNCaeTuMr4%2BfmptLRUP/30k7Nok05d6/XJJ59o69atysvL09GjR%2BX479OHjh07JulUolaZyoqgM11%2B%2BeWVtp9%2BLt3pAq2yZWVlZWctq6io0GeffabNmzcrLy9Pdrvd2bfc3Nzf7WuzZs0qba9Tp44kqbS09PcOBQBcULQBXqZhw4ZKSEhQQkKCpFMX7n/55ZdatWqVNm/erB07duiRRx7Re%2B%2B9p1q1ajm3CwkJOec%2Bg4KCJJ26WUFyTaTWr19/3j6dLsQkKScnR6NGjXIWgBfq9/opyeWYLmb5mfbt26dRo0adNW1aVTzAGIA7UbQBXu6qq67SVVddpYSEBC1dulTPP/%2B8cnNztWbNGsXHxzvX%2B70px9NJ0ul1Tk8v1qpVS1lZWVX%2BtVgnTpzQY489pn379qlFixYaOXKkunbtqtDQUGcxFRsbq19%2B%2BeWc%2B7iUv4Jr9OjR%2Bv7779WwYUMlJyerZ8%2BeatiwobMYGzp0qP7zn/9csv4AqNko2oAaZPjw4XrjjTe0e/du7dixw2XZ6V9zVZnTy04nbvXr15d0anqvoKDA5UL737Nhwwbt27dPPj4%2BWrhwoa644oqz1qkuD/3Nzs5WVlaWJCk1NVVdu3Y9a53q0lcANQN3jwIWt27dOj399NNauHBhldY/Pb14%2Brqq03788cdK1y8vL3cmX6ev0Trz%2BrffFn%2B/5/Tdp82bN6%2B0YNu9e7cOHTpU5f2Z6eeff5Yk5/PgfquoqOh3HzoMAO5G0QZYXFZWltLT07Vw4ULt27fvd9c9cOCAcnJyJOmsOyz37Nmjbdu2nbXN119/7fxNAh06dJB06qaH6OhoSdJbb71V6Xvl5uYqPj7e%2BWgQ6X9J3cmTJyvdZsGCBc4///ahvJdaYGCgpFNTw5XdMPCXv/zFmbRVdgMDALgbRRtgcffff7/Cw8N17NgxDR8%2BvNJfA%2BVwOPTFF1/ooYceUmlpqa699lp1797dZZ2goCBNnjzZJemy2%2B3605/%2BJOnUozYuu%2Bwy57IRI0ZIkjIyMvT666877/6UTqVUo0ePVk5Ojvbv3%2B9sb9OmjSTp119/VWZmprO9uLhYM2bMUFZWlrMw3LVr10WPiTtERkbK19dX5eXl%2Btvf/uZsLysr02uvvab09HT17t1b0v/uIgUAM3FNG2BxoaGhWrx4sZKTk/XTTz9p2LBhCg8P11VXXaV69erJbrdr7969zmeJXXfddUpLSzvrxoPY2Fjt2bNHffr0UUxMjGrVqqUtW7aosLBQgYGBevrpp13Wv%2BWWW/TII49o4cKFmjlzppYtW6arrrpKBQUF2rZtm8rLy9W2bVuNHz/euU2HDh3Uo0cPffrpp0pOTlb79u1Vp04dbd26VX5%2BflqyZInef/99bd68WcuWLdMPP/ygxx57TJ06dTJ/IH%2BjcePGuueee5Senq6ZM2dq9erVql%2B/vrZt26aioiKlpaVp7969%2Buijj7R27VoNHTpUQ4YM0R/%2B8IdL3lcANQNFG%2BAFIiMj9d577%2Bn999/Xhx9%2BqJycHH333XcqKSmRv7%2B/GjVqpK5du6p///7q1atXpXdg2mw2LV68WC%2B//LI%2B%2BOAD7du3TwEBAYqLi1NKSorLc9ZOGzdunLp166Y33nhD3377rTZu3Kg6deqoXbt26t%2B/v4YMGXLWYy/mzp2r1NRUrV%2B/Xlu3blVERITi4uI0YsQINW/eXI0bN9b333%2Bvb775Rjt27Likd4v%2B1uTJkxUSEqKMjAxt375dDRo0UOfOnfXoo48qKipKJ06c0JdffqkNGzZox44dLmkjALibj4OfMkCNNnHiRL377ru68847nVOhAIDqh2vaAAAALICiDQAAwAIo2gAAACyAog0AANRY//73v9WtWzelpKT87noVFRWaM2eO%2BvTpo06dOunBBx/Unj17nMvtdrvGjh2rbt26qUePHnrqqafc/ltTuBEBAADUSIsWLdLbb7%2BtsLAwNW7cWHPmzDnnun/961%2B1ZMkSLVq0SI0aNdKcOXP01VdfadWqVfLx8VFycrJKSko0c%2BZMlZaWasyYMWrXrp0mT57stv6StAEAgBqpTp06evvttyv9tXq/lZ6ersTERLVs2VKBgYFKSUnRzp07tWXLFh08eFCZmZlKSUlRWFiYGjVqpJEjR2rFihWV/kaVi8Vz2gAAgGXl5eU5Hx5%2BWnh4uCIiIs677bBhw6r0HidOnNCPP/6otm3bOtsCAwN1xRVXKCsrS8eOHZOvr69at27tXB4dHa3jx49r165dLu1GULSZxYMPBDXFlVdKO3ZIV18tedGv7HFUcHWAFRw8eMzTXXArm81HYWEBOny4SBVe9HfQ3z/I011wGx8fKSBAKiqSvO0ioiBPnSaTPhfT581TWlqaS1tSUpKSk5Pd9h5HjhyRw%2BFQSEiIS3tISIgKCgpUv359BQYGujwM/PS6BQUFbusHRRuqpn59ydf31FdUaz4%2B3vch421sNh/5%2BPjIZvPxqqLNm/j4/O/Fv6fqbdCgQYqNjXVpCw8PN%2BW9fu82gEtxiwBFGwAAMJ/NnMvoIyIiqjQVakT9%2BvVls9lkt9td2u12uxo0aKCwsDAVFhaqvLxcvr6%2BzmWS1KBBA7f1gxsRAACA%2BWw2c16XQJ06dXT11VcrOzvb2Xb06FHt3r1b11xzjaKiouRwOJSTk%2BNcnpWVpeDg4Ep/b/PFomgDAAD4jQMHDqhfv37OZ7ElJCRo2bJl2rlzpwoLC5WamqqoqCjFxMQoLCxMcXFxmjt3rg4fPqxff/1VCxYs0D333CM/P/dNajI9CgAAzHeJUrELERMTI0kqKyuTJGVmZko6lZKVlpYqNzdXJSUlkqTBgwcrPz9fQ4cOVVFRkTp37uxyA8T06dP1zDPPqE%2BfPqpVq5Zuu%2B228z6w90LxcF2zeNvdox06SN98I3XsKG3e7OneuI033j3qjRdOe9vdo35%2BNoWGBqigoEhlZRWe7o7beNPdozbb/%2B4erfCeUyTJg3eP1qljzn5PnjRnv9UQSRsAADBfNUzarIaiDQAAmI%2BizTBGEAAAwAJI2gAAgPlI2gxjBAEAACyApA0AAJiPpM0wijYAAGA%2BijbDGEEAAAALIGkDAADmI2kzjBEEAACwAJI2AABgPpI2wyjaAACA%2BSjaDGMEAQAALICkDQAAmI%2BkzTBGEAAAwAJI2gAAgPlI2gyjaAMAAOajaDOMEQQAALAAkjYAAGA%2BkjbDGEEAAAALIGkDAADmI2kzjKINAACYj6LNMEYQAADAAkjaAACA%2BUjaDGMEAQAALICkDQAAmI%2BkzTCKNgAAYD6KNsMYQQAAAAsgaQMAAOYjaTOMEQQAALAAkjYAAGA%2BkjbDKNoAAID5KNoMYwQBAAAsgKQNAACYj6TNMEYQAADAAkjaAACA%2BUjaDKNoAwAA5qNoM4wRBAAAsACSNgAAYD6SNsMo2gAAgPko2gxjBAEAACyApA0AAJiPpM0wijYAAFBj/fLLL5o2bZq2bNmievXqqX///ho3bpxsvykyH3jgAX311VcubWVlZRo1apSSkpI0dOhQffPNNy7bXXnllVq9erXb%2BkrRBgAAzFdNk7bk5GRFR0crMzNThw4d0qOPPqqGDRvq/vvvd1lv8eLFLt8fPXpU/fv3V9%2B%2BfZ1tzz77rO666y7T%2Blo9R/ASGj58uP70pz%2B5tC1YsECDBw/2UI8AAPBCNps5LwOysrKUk5Oj8ePHKygoSC1atFBiYqLS09PPu%2B3cuXPVt29ftW7d2lAfLkSNT9oGDBig2bNna8KECc5I84MPPtC9995b5X3k5eUpPz/fpS38yisVUb%2B%2BW/vqUW3auH4FLiE/P%2B/6/6Wvr83lq7eopkHKRTl9LN50TJJUUeHpHrhfpZ/B4eGKiIg477bZ2dlq2rSpQkJCnG3R0dHKzc1VYWGhAgMDK93u559/1sqVK5WZmenSvmbNGv35z3/W/v371b59e02fPl3Nmze/iKOqXI0v2m655RZNnz5dX375pbp27ao9e/Zo586d%2BsMf/lDlfaSnpystLc2lLWnMGCWPGePu7nre3/7m6R64lY%2BnO2ASHy87sNDQAE93wRTBwXU93QWcR10vO0XHjnnwzU2qgCv9DE5KUnJy8nm3tdvtCg4Odmk7XcAVFBScs2hbuHCh7r77boWFhTnbWrZsqbp16yo1NVUVFRWaMWOGHnroIWVkZKh27doXeliVqvFFW0BAgG6%2B%2BWatXr1aXbt21QcffKDu3bu7nIjzGTRokGJjY13awm%2B/XVq61N3d9Zw2bU4VbEOGSDk5nu6N2zi%2B/sbTXXA7Hx/J4fB0L9zLbi/ydBfcytfXpuDgujp6tFjl5d4TfdSu7T3Ftc12qmArLvbOdMqbVPoZHB5e5e0dF/gD0263a9WqVfrnP//p0j516lSX76dPn67OnTvr66%2B/VteuXS/oPc6lxhdt0qkp0tGjR2vatGlat26d7rvvvgvaPiIi4uwYNjfXjT2sRnJypM2bPd0L1DBlZd75qVleXuFVx%2BbnhZ8oFRUUbW5jUtJW6WdwFYWFhclut7u02e12%2Bfj4nDO8Wb9%2Bva688ko1a9bsd/cdGBiokJAQHThw4KL6Vhkvm62/OF27dlVAQID%2B8Y9/aMeOHerTp4%2BnuwQAgHephjcitGvXTvv379fhw4edbVlZWWrVqpUCAipPjtevX6/u3bu7tBUWFmrq1KkuBdrhw4d1%2BPDh8xZ3F4KiTZLNZtPtt9%2Bu2bNnq0%2BfPqrrbRcxAACAs7Rt21YxMTGaNWuWCgsLtXPnTi1ZskQJCQmSpH79%2BmnTpk0u22zbtk2XX365S1tgYKC2bNmiGTNmyG6368iRI5o2bZpat26tDh06uK2/FG3/NWDAABUWFur222/3dFcAAPA%2B1TBpk6R58%2BYpLy9P3bt317BhwzRgwAANGTJEkpSbm6vjx4%2B7rJ%2Bfn6%2BGDRuetZ8FCxbI4XAoLi5OvXr1UmlpqRYuXHjWQ3qN8MIrEC7OoUOH1LRp07MiTwAA4L0aN26sRYsWVbrshx9%2BOKtt69atla572WWXnXUXq7tRtOnUM16ef/55Pfjgg26tiAEAwH/x%2BWpYjR/B1157TX/4wx/UqVMn5xw2AABws2o6PWolNT5pe/TRR/Xoo496uhsAAAC/q8YXbQAA4BKoYamYGRhBAAAACyBpAwAA5iNpM4yiDQAAmI%2BizTBGEAAAwAJI2gAAgPlI2gxjBAEAACyApA0AAJiPpM0wijYAAGA%2BijbDGEEAAAALIGkDAADmI2kzjBEEAACwAJI2AABgPpI2wyjaAACA%2BSjaDGMEAQAALICkDQAAmI%2BkzTBGEAAAwAJI2gAAgPlI2gyjaAMAAOajaDOMEQQAALAAkjYAAGA%2BkjbDGEEAAAALIGkDAADmI2kzjKINAACYj6LNMEYQAADAAkjaAACA%2BUjaDKNoAwAA5qNoM4wRBAAAsACSNgAAYD6SNsMYQQAAAAsgaQMAAOYjaTOMog0AAJiPos0wRhAAAMACSNoAAID5SNoMYwQBAAAsgKQNAACYj6TNMIo2AABgPoo2wxhBAAAACyBpAwAA5iNpM4wRBAAANdYvv/yiRx55RJ07d1bv3r310ksvqaKi4qz15s%2Bfr6ioKMXExLi8Dh48KEk6efKknn76afXs2VOdO3fW6NGjVVBQ4Na%2BUrQBAADz2WzmvAxKTk5Wo0aNlJmZqSVLligzM1NLly6tdN34%2BHhlZWW5vBo2bChJmjNnjrKzs5Wenq61a9fK4XBo0qRJhvt3Joo2AABgvmpYtGVlZSknJ0fjx49XUFCQWrRoocTERKWnp1/QfsrKyvT2229r5MiRatKkierXr6%2BxY8fq448/1oEDBwz18Uxc02YSR4XD011wOx9Jjq%2B/8XQ33MrH5uPpLrhXhw7SN9/I57qO0ubNnu6N%2B%2BQd9XQPUAVBtiJPd8F9bDZJdRVgK5Z09lSZtQV4ugNulZeXp/z8fJe28PBwRUREnHfb7OxsNW3aVCEhIc626Oho5ebmqrCwUIGBgS7r//DDDxo8eLC2b9%2BuJk2aaNKkSerRo4d2796tY8eOKTo62rluy5Yt5e/vr%2BzsbDVq1MjgUZ5C0QYAAMxn0o0I6enpSktLc2lLSkpScnLyebe12%2B0KDg52aTtdwBUUFLgUbY0bN1azZs00btw4RUREKD09XSNGjNDq1atlt9sl6ax9BQcHu/W6Noo2AABgWYMGDVJsbKxLW3h4eJW3dziqNjM2cOBADRw40Pl9YmKi3n//fa1evVo9e/a8oH1dLIo2AABgPpOStoiIiCpNhVYmLCzMmZKdZrfb5ePjo7CwsPNu37RpU%2BXl5TnXtdvtCgj43/TzkSNH1KBBg4vqW2W4EQEAAJivGt6I0K5dO%2B3fv1%2BHDx92tmVlZalVq1YuxZckvfzyy9q4caNL286dO9WsWTM1a9ZMISEhys7Odi7bvn27SkpK1K5dO0N9PBNFGwAAqJHatm2rmJgYzZo1S4WFhdq5c6eWLFmihIQESVK/fv20adMmSadStGnTpmnXrl06efKkFi9erN27d%2BvOO%2B%2BUr6%2Bv7r33Xr366qvav3%2B/CgoKNHv2bPXt29f5SBB3YHoUAACYr5r%2BRoR58%2BZpypQp6t69uwIDAzV48GANGTJEkpSbm6vjx49LksaNGyfp1LVsdrtdrVq10uuvv67GjRtLkkaPHq2ioiLFx8errKxMvXv31tSpU93aVx%2BH2VfN1VDeOKo%2BPt53XN76yA919K5HfuR72SM//PxsCg0NUEFBkcrKvOdxEuH1queH8kWx2aS6daXiYqmSp%2BNbWoCHHvkxc6Y5%2B3XzA2yrM5I2AABgvmqatFkJRRsAADAfRZthjCAAAIAFkLQBAADzkbQZxggCAABYAEkbAAAwH0mbYRRtAADAfBRthjGCAAAAFkDSBgAAzEfSZhgjCAAAYAEkbQAAwHwkbYZRtAEAAPNRtBnGCAIAAFgASRsAADAfSZthjCAAAIAFkLQBAADzkbQZRtEGAADMR9FmGCMIAABgASRtAADAfCRthjGCAAAAFkDSBgAAzEfSZhhFGwAAMB9Fm2GMIAAAgAWQtAEAAPORtBlG0QYAAMxH0WYYIwgAAGABJG0AAMB8JG2GMYIAAAAWQNIGAADMR9JmGEUbAAAwH0WbYYwgAACABZC0AQAA85G0GcYIAgAAWABJGwAAMB9Jm2EUbQAAwHwUbYYxggAAABZA0gYAAMxH0mYYIwgAAGABJG0AAMB8JG2GUbQBAADzUbQZxggCAABYQLUp2vbu3avWrVtr586dl/R9hw4dqtTU1Ev6ngAA1Dg2mzmvGqRmHa2kPXv26F//%2BpenuwEAAKqBX375RY888og6d%2B6s3r1766WXXlJFRUWl6y5fvlxxcXHq0KGD4uPjlZmZ6Vw2ceJEtW3bVjExMc7X9ddf79a%2B1rhr2j744ANt3bpV/fr183RXAACoOappKpacnKzo6GhlZmbq0KFDevTRR9WwYUPdf//9LuutXbtWs2bN0muvvaZrrrlGK1eu1NixY/XPf/5TzZo1kyQ99thjSk5ONq2v1bJos9vtmjFjhr744gsVFRWpS5cumjp1qho1aqS9e/eqT58%2BWrx4sV566SX99NNPatOmjVJTU3X55ZdLkl5%2B%2BWUtWbJEfn5%2BevTRR/XJJ5%2BoY8eOqlevnnMqNDMzU99%2B%2B60kqby8XE8//bQyMjJUp04dTZkyRf37969yf/Py8pSfn%2B/S1rBhuCIiItw0IjBNhw6e7oF7tWnj%2BtVL%2BPlVzx/2F8vX1%2Bby1WtU0w/li%2BLj87%2Bv3nRc50iQLgmTxrGyz%2BDw8Kp9BmdlZSknJ0dLlixRUFCQgoKClJiYqKVLl55VtJ04cUKPP/64rrvuOknSwIEDlZqaqm%2B//dZZtJmtWhZtEydOlJ%2Bfn95//335%2BvrqmWee0aRJk7R48WLnOsuWLdNrr72mOnXqaNiwYfrzn/%2BsqVOnat26dXr11Ve1ePFitW3bVjNnzlR2drY6duyoBx98UDt27NDJkyc1Z84c574yMjL0/PPPa/LkyUpLS9PUqVN1yy23yM%2BvasOTnp6utLQ0l7ZRo5I0erR51bannP455jW%2B%2BcbTPTDH3/7m6R64VainO2CS4OC6nu4Czsff39M9cK%2BiIk/3wO0q%2BwxOSkqqUuKVnZ2tpk2bKiQkxNkWHR2t3NxcFRYWKjAw0NkeHx/vsu3Ro0dVVFSkRo0aOdu%2B%2BOILrV%2B/Xj///LNatmypqVOnql27dhd7aGepdkXb4cOH9dFHH2nNmjXOQRw/frx69erlUkknJCQ4B6pHjx7KysqSJH3yySfq0aOHcx55woQJWrVq1e%2B%2BZ8eOHXXjjTdKkvr166fXXntNhw8frnJSNmjQIMXGxrq0NWwYLoejSptbho%2BPvO%2BYruvo6S64V5s2pwq2IUOknBxP98ZtCtb/29NdcCtfX5uCg%2Bvq6NFilZd7MPlws1B/L0qkfHxOFWwnTnjfDz5PMSlpq%2BwzODw8vErb2u12BQcHu7Sdrj0KCgpcirYzORwOTZ48We1CfuqZAAAgAElEQVTbt9cNN9wgSWrWrJlsNpvGjBmjgIAApaWl6YEHHtDatWsVGuqe/3pWu6KtVq1akqQBAwa4tPv6%2Bmr//v0KCwuTJOdUqCTVrVtXJ0%2BelCTl5%2BerefPmzmVBQUFq0aLF777nmfuqU6eOJKmkpKTKfY6IiDirwOPfuEVs3uzpHpgjJ8erjq2szHsKmzOVl1d417F50aE4CwyHw7NTijivyj6DL4TjAj%2BwS0tLNXHiRP34449atmyZs33UqFEu6z3xxBPKyMhQZmamBg4ceNH9O1O1K9r8/xtFb9iwodLKdO/evZIkn3PM01VUVJw1rWk7T3V/rn0BAAA3qYbXBoaFhclut7u02e12%2Bfj4OEOiM504cUIjR45UcXGx3nzzzd9N0Hx9fdWkSRPl5eW5rb/VbwR1qsj64YcfnN%2BXlpbqwIEDVdq2QYMG2rdvn/P7wsJC5ebmur2PAADgAlTD57S1a9dO%2B/fv1%2BHDh51tWVlZatWqlQICAlzWdTgcSklJkZ%2Bfn15//XWXgs3hcGjmzJnKOeOylJKSEu3evdutNylUu6KtVq1a6t%2B/v1JTU/Xrr7/qxIkTmj17th544IEqRZhdunTRhg0b9N133%2BnEiRN68cUXnemddGr6c//%2B/Tp69KjKysrMPBQAAFCNnX6u2qxZs1RYWKidO3dqyZIlSkhIkHTqOvdNmzZJkt577z39%2BOOP%2Br//%2Bz/npVSn%2Bfj4aO/evZo2bZoOHDigoqIipaamqlatWrr55pvd1t9qV7RJ0pQpU3TFFVfo1ltv1Y033qgff/xRL7/8cpWmMe%2B44w7dfffdGjZsmOLi4tS%2BfXs1b97cue3tt9%2Bu3Nxc9e7d262RJQAA%2BB3VMGmTpHnz5ikvL0/du3fXsGHDNGDAAA0ZMkSSlJubq%2BPHj0uSVqxYoV9%2B%2BUU33HCDywN0J0%2BeLEl67rnn1KJFC911113q1q2btm3bpqVLl6pevXqG%2B3iaj%2BNCr8CzgJKSEtWuXdv5fe/evTVy5Ei3XQhYFd43ql5696jNy65n7NDh1GNMOnb0qhsR8vOOeroLbuXnZ1NoaIAKCoq86kaE8HrVMge4ODabVLeuVFzsfTci/Gba75LZuNGc/Xbtas5%2BqyEv%2Bhd2yldffaVOnTrpu%2B%2B%2BU3l5ud555x3l5%2Beraw06qQAAVDvVNGmzkmp396hRnTp1UkpKisaOHavDhw%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%2BfP14osv6vPPP1fv3r01YsQIHT9%2BXJI0Z84cZWdnKz09XWvXrpXD4dCkSZMM9%2B9MFG0AAMB8JhVteXl5ys7Odnnl5eVVqUtZWVnKycnR%2BPHjFRQUpBYtWigxMVHp6elnrZuenq677rpL7du3l7%2B/vx566CFJ0kcffaSysjK9/fbbGjlypJo0aaL69etr7Nix%2Bvjjj3XgwAG3DSHXtJnk4MFjnu6CW/n52RQaGiC7vUhlZZXHxpaUd9TTPXArPz%2BbQiUVrP%2B3d50nWEL%2Bce/5O%2BfnJ4XWlQpOVHjVv6Xw8CBPd8Ht0tPTlZaW5tKWlJSk5OTk826bnZ2tpk2bKiQkxNkWHR2t3NxcFRYWKjAw0GXd/v37O7%2B32WyKiopSVlaWoqKidOzYMUVHRzuXt2zZUv7%2B/srOzlajRo2MHKITRRsAADCdQz6m7HfQoEGKjY11aQsPD6/Stna7XcHBwS5tpwu4goICl6LNbre7FHen1y0oKJDdbpeks/YVHBzs1uvaKNoAAIBlRUREKCIi4qK3dzgcblv3QvZ1MbimDQAAmK6iwpyXEWFhYc6U7DS73S4fHx%2BFhYW5tIeGhla6blhYmHPd3y4/cuSIGjRoYKyTZ6BoAwAApquORVu7du20f/9%2BHT582NmWlZWlVq1aKSAg4Kx1s7Oznd%2BXl5fr%2B%2B%2B/V/v27dWsWTOFhIS4LN%2B%2BfbtKSkrUrl07Y508A0UbAACokdq2bauYmBjNmjVLhYWF2rlzp5YsWaKEhARJUr9%2B/bRp0yZJUkJCglauXKlvv/1WxcXFeuWVV1S7dm316tVLvr6%2Buvfee/Xqq69q//79Kigo0OzZs9W3b181bNjQbf3lmjYAAGA6o6mYWebNm6cpU6aoe/fuCgwM1ODBgzVkyBBJUm5urvM5bD179tTjjz%2BusWPH6tChQ4qJidHChQvl7%2B8vSRo9erSKiooUHx%2BvsrIy9e7dW1OnTnVrX30cZl81V0Pl53vnIz8KCrzskR9ehvNkDZyn6s9bz5EnH/lx8qQ5%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%2BPyOnjwYJX7U2OvaSssLFT79u1ls9XYuhUAgEvGiqnYlClTVFJSooyMDJWWlmrMmDFKTU3V5MmTz1r3u%2B%2B%2B0/jx4zV79mz16tVLn332mUaNGqWrrrpK119/vSQpPj5ef/rTny66PzWyaLvvvvu0adMmff311/rggw8UGxurjIwMFRYWqnnz5nriiSd04403Vnl/eXl5ys/Pd2nz86uniIgId3fdY3x9bS5fUT1xnqyB81T9cY7cz6yirbLP4PDwcMOfwQcPHlRmZqbeffddhYWFSZJGjhypMWPG6Mknn1StWrVc1rfb7Xr00Ud18803S5JuuukmRUZGatOmTc6izagaWbS98cYbGjp0qNq3b6/WrVtr0aJFWr16tUJCQrRy5Uo9%2BeST%2BuSTT846IeeSnp6utLQ0l7ZRo0Zp9OjRZnTfo4KD63q6C6gCzpM1cJ6qP85R9VfZZ3BSUpKSk5MN7Xfbtm3y9fVV69atnW3R0dE6fvy4du3a5dIuST179lTPnj2d35eVlSk/P1%2BNGjVytv3www8aPHiwtm/friZNmmjSpEnq0aNHlftUI4u2Mx09elR%2Bfn6qW7eufH19dffdd%2BvOO%2B%2B8oGnTQYMGKTY21qXNz6%2BeCgqK3N1dj/H1tSk4uK6OHi1WebkFM%2B4agvNkDZyn6s9bz1FoaIDH3tuspK2yz%2BDw8HDD%2B7Xb7QoMDJSPj4%2BzLSQkRJJUUFBw3u1TU1NVr1495w2PjRs3VrNmzTRu3DhFREQoPT1dI0aM0OrVq3XVVVdVqU81vmi79dZbtWrVKvXs2VPdu3dXr169dOutt15Q0RYREXFWDJuff0xlZd7zD/208vIKrzwub8N5sgbOU/XHOar%2BKvsMrqpVq1ZpwoQJlS5LSUmRw%2BG44H06HA6lpqYqIyNDy5YtU506dSRJAwcO1MCBA53rJSYm6v3339fq1as1duzYKu27xhdt9evX11tvvaVvvvlGH330kebNm6fly5frzTfflJ9fjR8eAADcojreiBAfH6/4%2BPhKl3322WcqLCxUeXm5fH19JZ1K3ySpQYMGlW5TUVGhSZMm6bvvvtPy5cvVrFmz333/pk2bKi8vr8r9rfFXWJ48eVLFxcXq2LGjxo0bp4yMDG3fvl05OTme7hoAAF7Das9pi4qKksPhcKkHsrKyFBwcrCuvvLLSbZ5//nnt2LGj0oLt5Zdf1saNG13adu7ced7C7kw1vmh77rnn9OSTT%2Brw4cNyOBzKzs5WRUWFLrvsMk93DQAAeEhYWJji4uI0d%2B5cHT58WL/%2B%2BqsWLFige%2B65xzkTN3z4cK1Zs0aS9PXXX2v16tVauHCh6tevf9b%2B7Ha7pk2bpl27dunkyZNavHixdu/erTvvvLPKfarx83/jxo3TM888o7i4OJWVlemKK67QrFmznLf3AgAA46rj9Oj5TJ8%2BXc8884z69OmjWrVq6bbbblNKSopz%2BZ49e3TkyBFJ0ooVK3Ts2DH17t3bZR%2BdOnXS4sWLNW7cOEmnrmWz2%2B1q1aqVXn/9dTVu3LjK/fFxXMxVdjiv/Pxjnu6CW/n52RQaGqCCgiIuyq3GOE/WwHmq/rz1HIWHB3nsvf/5T3P2%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%2B%2BKD27NlzQf2haAMAAKazYtE2ZcoUFRcXKyMjQytWrNDOnTuVmpp6zvU7deqkrKwsl9c111wjSXrzzTf13nvvaeHChfroo4/UokULjRo1Sg6Ho8r9oWgDAAD4jYMHDyozM1MpKSkKCwtTo0aNNHLkSK1YsUKlpaUXvL/09HQlJiaqZcuWCgwMVEpKinbu3KktW7ZUeR/ciAAAAExnViqWl5en/Px8l7bw8HBFREQY2u%2B2bdvk6%2Bur1q1bO9uio6N1/Phx7dq1y6X9tP379%2Bv%2B%2B%2B/X1q1bFRwcrNGjRys%2BPl4nTpzQjz/%2BqLZt2zrXDQwM1BVXXKGsrCxde%2B21VeoTRZtJ/P2DPN0Ft7L9N5OtXTtAfl70tybIVuTpLrjXf09UqL9N8qI7tfKPe9HBeLHwiGBPd8F9OnSQvvlGoX1ulDZv9nRv3OsCpuOsID09XWlpaS5tSUlJSk5ONrRfu92uwMBA%2Bfj4ONtCQkIkSQUFBWetHxYWphYtWujxxx9Xq1attG7dOk2YMEERERG66qqr5HA4nNufub/K9nUuXvTxCwAAqiuzkrZBgwYpNjbWpS08PLxK265atUoTJkyodFlKSsoFXW/Wq1cv9erVy/n9rbfeqnXr1umdd97R%2BPHjJemC9lcZijYAAGA6s4q2iIiIi54KjY%2BPV3x8fKXLPvvsMxUWFqq8vFy%2Bvr6STqVvktSgQYMq7b9p06baunWr6tevL5vN5tz%2BNLvdXuV9SdyIAAAAcJaoqCg5HA7l5OQ427KyshQcHKwrr7zyrPWXL1%2BuNWvWuLTt3LlTzZo1U506dXT11VcrOzvbuezo0aPavXu38%2B7SqqBoAwAAprPaIz/CwsIUFxenuXPn6vDhw/r111%2B1YMEC3XPPPfL778Xdw4cPdxZqJSUlevbZZ5WVlaXS0lJlZGRow4YNGjx4sCQpISFBy5Yt086dO1VYWKjU1FRFRUUpJiamyn1iehQAAKAS06dP1zPPPKM%2BffqoVq1auu2225SSkuJcvmfPHh05ckSSNGzYMBUVFWnMmDHKz8/X5ZdfrgULFqhdu3aSpMGDBys/P19Dhw5VUVGROnfufNYNFOfj4zB6VRwqdeyYp3vgXjabFBAgFRV51%2B%2BP88q7R%2BvWlYqLvepEedvdo35%2BNoWGBqigoEhlZd5zbN5496g6duTuUTeZNMmc/c6cac5%2BqyOSNgAAYDov%2Bn%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%2BrKxMAwYMUL9%2B/Zxtjz32mJKTky%2B6P0yPAgAA/MbBgweVmZmplJQUhYWFqVGjRho5cqRWrFih0tLS826/dOlSXXbZZbrpppvc1qcLStr27t2rPn36aM2aNXr44Yf12GOPad26dfrqq6/UoEEDTZ06VT169JAkffjhh3rhhReUl5enm266SQ0bNqzy%2B0ycOFEBAQHy8/PTu%2B%2B%2BK5vNpgcffFAPP/ywJKl169ZatGiRevbsKUlavny5Fi1apA8//NDZx9dee00vvPCC9u3bp/79%2B%2Buxxx7ThAkTlJOTo5iYGKWlpSkkJESS5HA4NGPGDK1cuVKBgYF68MEHL2gqNy8vT/n5%2BS5t9eqFKyIiosr7qO5sNtevXsPbDsjH539fvejY/LzslilfX5vLV6/RoYOne%2BA%2Bbdq4fvUWmzd77K3NStoq%2BwwODzf%2BGbxt2zb5%2BvqqdevWzrbo6GgdP35cu3btcmn/raNHj%2BrVV1/V3/72N5f2L774QuvXr9fPP/%2Bsli1baurUqWrXrl2V%2B2ToR%2BFf/vIXvfjii2rTpo2mTp2q559/XmvWrNHRo0eVkpKi8ePHa9CgQdq4caPGjRunqKioKu87IyNDEydO1GeffabVq1drypQpio%2BPr/JJWLlypd566y3l5OTovvvu088//6yXXnpJ/v7%2BuvPOO7VixQo98MADkqQNGzboiSee0Oeff66PP/5YycnJat%2B%2Bva655poqvVd6errS0tJc2kaNStLo0RcfgVZXdet6ugfu5nUHdIq/v6d74FahXnqagoO97MC%2B%2BcbTPXC/33zoWt7p/9h5kco%2Bg5OSkgxNQ0qnrmcLDAyUzxljdjrsKSgo%2BN1t33jjDXXq1ElXX321s61Zs2ay2WwaM2aMAgIClJaWpgceeEBr165VaGholfpkqGjr3bu3s7CJi4vTypUrVVFRoU8//VT16tXT//t//082m0033XSTrr/%2BehUVFVV535dffrnuvPNOSVL//v31xz/%2BUT/99FOVi7a7775bQUFB6tSpk4KCgtS9e3c1a9ZMknTNNdfop59%2Bcq4bERGhhIQESdItt9yiqKgobdiwocpF26BBgxQbG%2BvSVq9euC7gcKs9m%2B1UwVZcbM3rEs4lwFbs6S64l4/PqYLtxAnJ4fB0b9ym4IQX/aXTqYQtOLiujh4tVnm59xxbaJ8bPd0F92nT5lTBNmSIlJPj6d54BbM%2BOyr7DA4PD6/StqtWrdKECRMqXZaSkiLHRfwcLS8v15tvvqlZs2a5tI8aNcrl%2ByeeeEIZGRnKzMzUwIEDq7RvQ0Xb5Zdf7vyzv7%2B/ysvLVVpaql9//VVNmjSR7YzpmRYtWig7O/ui9l33v/HOue7YqEyTJk2cf65Tp44aNWrk8n1JSYnz%2B1atWrls27x5cx04cKDK7xUREXFWMXnsmHcVN6ddirt1Li2vOpj/TYk6HF51osrKvOdYzlReXuFdx%2BbBqTfT5OR453F5gFk/kir7DK6q%2BPh4xcfHV7rss88%2BU2FhocrLy%2BXr6yvpVPomSQ0aNDjnPr/66iuVlJSc985QX19fNWnSRHl5eVXur6ELKmznuGampKRE5eXlLm0VF3i2zrXvylS2b5/fRMC/t7/fLnM4HKpTp06V3x8AAHiXqKgoORwO5ZyRtGZlZSk4OFhXXnnlObdbv369unTpIr8zLsZ1OByaOXOmy75KSkq0e/du5yxgVZhyFWxERIQOHDjgEivu3LnTbfuvXbu2S%2Bq2e/duQ/v77XPndu/e7VU3EQAA4GlWe05bWFiY4uLiNHfuXB0%2BfFi//vqrFixYoHvuucdZkA0fPlxr1qxx2W7btm0us4XSqSBp7969mjZtmg4cOKCioiKlpqaqVq1auvnmm6vcJ1OKtm7duqmwsFB///vfVVJSoszMTG3ZssVt%2B2/RooUyMzNVVlamrKwsffzxx4b2t3fvXr377rsqLS3V%2BvXr9cMPP6hv377u6SwAALCk6dOnKygoSH369NEdd9yha665xuWB/Hv27NGRI0dctsnPz6/0iRnPPfecWrRoobvuukvdunXTtm3btHTpUtWrV6/K/THlRvrGjRtr1qxZSk1N1QsvvKCePXtqyJAh2uym6wL%2B%2BMc/6umnn9b111%2Bv66%2B/Xg888IBee%2B21i95fXFyctmzZohkzZigoKEjTp0//3egTAABcGCteZhsUFKTZs2efc/mHH354VtvatWsrXbd%2B/fqaOXOmof74OC7m1gic17Fjnu6Be9lsUkCAVFRkzX945xJk86JbfCWvvc03/7j3HIsk%2BfnZFBoaoIKCIq%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%2BrVq5WcnKwrrrjivOt%2B%2BOGHmj9/vl588UV9/vnn6t27t0aMGKHjx49LkubMmaPs7Gylp6dr7dq1cjgcmjRp0gX1h6INAACgEidPnlR6errat29/3nXT09N11113qX379vL399dDDz0kSfroo49UVlamt99%2BWyNHjlSTJk1Uv359jR07Vh9//LEOHDhQ5f5wIwIAADCdWalYXl6e8vPzXdrCw8MVERFheN8DBw6s8rrZ2dnq37%2B/83ubzaaoqChlZWUpKipKx44dU3R0tHN5y5Yt5e/vr%2BzsbDVq1KhK70HRZpKgIE/3wL3y8vK0eHG6Bg0a5JZ/CNVHgKc74FZ5eXlK//Ofve48hXvXaVJeXp7mz1/sdedJDoene%2BA2eXl5Sp8/X4P%2B9S/vOkceZNZfj/nz05WWlubSlpSUpOTkZHPe8BzsdrtCQkJc2kJCQlRQUCC73S5JCg4OdlkeHBx8Qde1UbShSvLz85WWlqbY2Fh%2BgFVjnCdr4Nn4MAMAAAy4SURBVDxVf5wj6xg0aJBiY2Nd2sLDw6u07apVqzRhwoRKl82cOVN33XXXBfXFcZ7K9HzLz4eiDQAAWFZERMRFF9bx8fGKj493Sz9CQ0OdidppdrtdV199tcLCwpzfBwT8b%2BrgyJEjatCgQZXfgxsRAAAADGrXrp2ys7Od35eXl%2Bv7779X%2B/bt1axZM4WEhLgs3759u0pKStSuXbsqvwdFGwAAwEXo16%2BfNm3aJElKSEjQypUr9e2336q4uFivvPKKateurV69esnX11f33nuvXn31Ve3fv18FBQWaPXu2%2Bvbtq4YNG1b5/ZgeRZWEh4crKSmpytcJwDM4T9bAear%2BOEeQpLi4OO3bt0/l5eWqqKhQTEyMJOlf//qXmjZtqtzcXOdz2Hr27KnHH39cY8eO1aFDhxQT8//bu9uYrKsGjuO/CxgP0nrABcmDSHOTWreCGIhQCAU5Q4FGRi6tLF2r0SZlBq0nR7JqYY6eY2bUC1tFOalFgIChS3BjOgppEqhBI7EUKBG88H4Bcnt5XT7tVo7/%2BH5e4Tn/F7/N6X6c8z/n/x998MEH8vb2liQ9%2BeST%2Bvvvv5WWlqYTJ04oMTFRL7300kXlsZ38f9%2BKAwAAwGXH9igAAIAFUNoAAAAsgNIGAABgAZQ2AAAAC6C0AQAAWAClDQAAwAIobQAAABZAaQMAALAAShsAAIAFUNrg0v33369NmzbpyJEjpqMAAADxGSucxdtvv62Kigq1trYqLi5OCxcu1B133CEvLy/T0eDC0NCQ/vzzTw0MDDjNBQYGGkiEM9ntdrm7u0sa/vvau3evJk2apOuuu85wMgBWQWnDOR08eFAVFRWqqKjQvn37dOedd2rBggWaM2eO6WgY8e233%2Brll19WT0%2BPw/jJkydls9nU3NxsKBlOqaur07PPPqu6ujqdOHFCDzzwgFpaWiRJhYWFSkxMNJwQp2tqalJtba26urrk5eWlgIAAJScnKzQ01HQ0jHOUNlyQgYEBffHFFyosLFRfX5%2BCgoK0fPlyZWVlmY427sXFxSkzM1Pz5s1zuRJ64403GkiF06Wnp2vp0qW65557tHnzZhUVFWnLli1qampSQUGBSktLTUfEiE8//VT5%2Bfm66aabFBISIkk6cOCAfvnlF61Zs0aZmZmGE2I88zAdAFe2H3/8UVu2bNH3338vX19fZWVlKT09Xd3d3SooKFBra6uee%2B450zHHtf7%2BfmVnZ8vDg3/OV6r9%2B/crIyNDklRTU6O7775bPj4%2BuvXWW9Xe3m42HBxs2LBBRUVFSk5OdhivqqrSK6%2B8QmmDURxEgEuvvvqqEhIS9Pjjj2twcFDr169XdXW1nn76aU2dOlWzZ8/Whx9%2BqK%2B//tp01HEvNTVV9fX1pmPgHHx9fdXT06P%2B/n5t3759dDv0r7/%2BomxfYXp7ezV37lyn8YSEBKdXEICxxv8WcKm5uVkrV65USkqKJkyY4PIZf39/LV%2B%2BfIyTQRp%2BD%2BqUCRMmKDc3VzNnzlRwcLBsNpvDszk5OWMdD2dYsGCBHnzwQbm7uyssLEwRERHq7%2B/Xiy%2B%2BqLi4ONPxcJq0tDR99dVXWrRokcN4WVmZFi5caCgVMIzSBpc2btx4Qc%2BtWLHi8gaBS42NjQ5/njx5srq7u9Xd3e0wfmaBw9hJTU1VWVmZJGnr1q3Kzs5WT0%2BPUlNTJUlubm669tprtWrVKpMxIempp54a/XloaEivvfaaNm7cqKlTp8pms6mtrU0dHR1KSUkxmBLgIAJgeadOiZ7Jbrfr0KFDuuGGGwykQkJCgiIjIzV58mRt2LBBy5YtO%2BuzrIaalZube8HPFhQUXMYkwLlR2gCLmzFjhnbv3u003tvbq6SkJDU0NBhIhZ07d%2Brjjz9WX1%2BfGhoaNGvWLJfP2Ww2lZSUjHE6AFbE9ihgUeXl5SovL9fg4KDD9s4pnZ2do5e5YuzFxMQoJiZGkrRkyRJ98sknhhPhQvT29urLL79Ue3u7jh8/7jBns9m0du1aQ8kAShtgWTfffLN%2B%2B%2B03fffdd/L09HSanzZtmssyh7FHYbOOJ554Qm1tbZo%2BfTpfgMEVh%2B1RwOLO974UgAsXGRmpyspKTZw40XQUwAkrbYDF/fPPP3rrrbdczrm5uSkgIECxsbF8gxS4ANOmTXPaFgWuFKy0ARa3bNky/fTTT%2Brv79eUKVPk5uamtrY2%2Bfj4KCQkZPQqkPXr1/ONS%2BA8Dh48qNzcXMXFxSkgIEBubo530KenpxtKBrDSBlheQkKCgoOD9cwzz%2Biqq66SJPX19en1119XRESEMjIyVFpaqnXr1lHagPN49913tWvXLrW0tLh8p43SBpNYaQMsLi4uTlVVVfL29nYYP378uObNm6fq6moNDQ0pKirK6VJeAI4iIiJUXFx81itaAJP49ihgcYODg/r555%2BdxltaWnTkyBFJUlNTk/z8/MY6GmA5gYGBCg8PNx0DcIntUcDiHnroIT388MO67bbbFBQUJA8PD3V2dqq2tlaZmZkaGBjQ0qVLtXLlStNRgSve6tWr9cILLygzM9PlO21hYWGGkgFsjwL/ChUVFdq2bZsOHTqkoaEhTZw4UTExMUpLS5PNZtOuXbvY7gEugKtVNpvNNvq5uObmZgOpgGGUNgAARnR0dJxzPigoaIySAM4obYDFdXZ2qri4WK2trS7vl9q0aZOBVACAS4132gCLy8nJ0bFjxxQfHy8fHx/TcQBLS0pKks1mO%2Bt8VVXVGKYBHFHaAItraWlRbW2trr76atNRAMtbsWKFw5/tdrsOHDigmpoaPfroo4ZSAcMobYDFTZkyRQMDA6ZjAP8KWVlZLsdTU1NVVFSke%2B%2B9d4wTAf/DO22Axe3YsUMlJSVavHixgoKCuKIAuAzsdrtmzZrFBdUwitIGWBxXFACXTl1dndNYf3%2B/Kisr1djYqPLycgOpgGGUNsDiuKIAuHRc/RLk6emp0NBQ5eXlKTY21kAqYBilDfiXOHz4sH7//XfdcsstpqMAAC4DDiIAFtfV1aW8vDxt375dHh4eampq0h9//KFHHnlE77zzjkJCQkxHBCylu7tb7e3t6u/vd5qLj483kAgYRmkDLG7NmjXy8/NTdXW17rrrLkmSn5%2Bf4uPjlZ%2Bfr/fff99wQsA6iouLtW7dOtntdqc53hGFaWyPAhYXFRWlbdu2ydfXVzNmzNDu3bslDb88nZCQoJ07dxpOCFhHbGysVq1apfnz58vb29t0HMABK22Axfn4%2BMjV715Hjx51uVoA4OyGhoaUlpYmd3d301EAJ27nfwTAlWz27NnKy8tTW1ubJKmnp0f19fXKzs7W3LlzzYYDLCYjI0NlZWWmYwAusT0KWNzRo0e1evVq1dTUjI65ublp/vz5ev7553XNNdeYCwdYTH5%2Bvr755hsFBwcrODjY6bLqN954w1AygNIGWNKpVbXTHT58WJ6envL09JQkeXl5SeKLCMDFyM3NPed8QUHBGCUBnFHaAAsKDw8f/eqBNHyq7Ux8EQG4fN577z099thjpmNgnKG0ARZ0vq8gnI4vIgCX3ukntYGxwulRwIIoYoBZrHfABE6PAgBwkVy9kgBcbpQ2AAAAC6C0AQAAWAClDQAAwAIobQAAXCQOIsAEShsAABdp0aJFpiNgHOKeNgDAuLZkyZILPg1aUlJymdMAZ8c9bQCAcS0iImL052PHjmnz5s2KiopSWFiYhoaGtG/fPu3Zs0eLFy82mBJgpQ0AgFE5OTlKT0/X7bff7jBeWVmpsrIyvfnmm4aSAZQ2AABGzZw5U/X19fLwcNyIGhwcVHR0tBobGw0lAziIAADAKH9/f3322WdO46Wlpbr%2B%2BusNJAL%2Bh5U2AABGbN26VTk5OfL19dWkSZNkt9vV1dWl3t5eFRYWKjk52XREjGOUNgAATtPb26sffvhBXV1dGhgYkL%2B/v%2BbMmaOAgADT0TDOUdoAAHBhYGDAaczT09NAEmAYV34AADBix44dWrt2rdrb22W3253mm5ubDaQChrHSBgDAiOTkZEVHRyslJUU%2BPj5O89HR0QZSAcMobQAAjIiMjFRDQ4PTlR/AlYArPwAAGBEdHa29e/eajgG4xEobAAAjPv/8c3300UdKTExUcHCw3Nwc1zbuu%2B8%2BQ8kAShsAAKOSkpLOOmez2VRVVTWGaQBHlDYAwLi2f/9%2BhYaGSpJ%2B/fVX2Wy2sz4bFhY2VrEAJ5Q2AMC4Nn36dO3Zs0eSFB4e7rK0nTx5UjabjSs/YBSlDQAwrnV2diowMFCS1NHRcc5ng4KCxiIS4BKlDQAAwAK48gMAAMACKG0AAAAWQGkDAACwAEobAACABVDaAAAALIDSBgAAYAGUNgAAAAv4L8Fz20yN3DDPAAAAAElFTkSuQmCC\" class=\"center-img\">\n",
       "</div>\n",
       "    <div class=\"row headerrow highlight\">\n",
       "        <h1>Sample</h1>\n",
       "    </div>\n",
       "    <div class=\"row variablerow\">\n",
       "    <div class=\"col-md-12\" style=\"overflow:scroll; width: 100%%; overflow-y: hidden;\">\n",
       "        <table border=\"1\" class=\"dataframe sample\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>y</th>\n",
       "      <th>x</th>\n",
       "      <th>length</th>\n",
       "      <th>fs</th>\n",
       "      <th>ind_numb</th>\n",
       "      <th>text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4</td>\n",
       "      <td>[22, 101, 83, 9, 27, 83, 63, 43, 38, 61, 45, 3...</td>\n",
       "      <td>30372</td>\n",
       "      <td>16000</td>\n",
       "      <td>3</td>\n",
       "      <td>a01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>6</td>\n",
       "      <td>[9, 3, 13, 13, 11, 15, 17, 15, 17, 24, 22, 22,...</td>\n",
       "      <td>25780</td>\n",
       "      <td>16000</td>\n",
       "      <td>3</td>\n",
       "      <td>a01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>[0, 1, -5, -1, -1, -1, -3, -1, -5, 0, -1, -9, ...</td>\n",
       "      <td>30045</td>\n",
       "      <td>16000</td>\n",
       "      <td>3</td>\n",
       "      <td>a01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>[23, 82, 88, -12, -114, -206, -112, -10, 59, 8...</td>\n",
       "      <td>32100</td>\n",
       "      <td>16000</td>\n",
       "      <td>3</td>\n",
       "      <td>a02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>6</td>\n",
       "      <td>[2, -10, -5, 0, 7, -5, 2, -10, -12, 0, -15, -1...</td>\n",
       "      <td>23037</td>\n",
       "      <td>16000</td>\n",
       "      <td>3</td>\n",
       "      <td>a02</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "    </div>\n",
       "</div>\n",
       "</div>"
      ],
      "text/plain": [
       "<pandas_profiling.ProfileReport at 0x7f1515a0dcf8>"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "profile_file = pandas_profiling.ProfileReport(data_complete)\n",
    "profile_file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    127\n",
       "1     81\n",
       "6     79\n",
       "4     71\n",
       "3     69\n",
       "5     62\n",
       "2     46\n",
       "Name: y, dtype: int64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_complete.y.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Some Comments\n",
    "Sampling rate is constant 16 kHz. But what appear as the main challenge with this data-set are:\n",
    "* The variable length of each audio-sample\n",
    "* The fact that the label 0 (Anger) has the double of samples than the rest (unbalanced data-set).\n",
    "\n",
    "### Strategies\n",
    "To address the variable length, we will proceed with the most _force brute straightforward_ techniques, **zero padding** and see what results we get."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "data_complete_pr = data_complete[(data_complete.length > data_complete.length.quantile(.05))&(data_complete.length < data_complete.length.quantile(.95))]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<meta charset=\"UTF-8\">\n",
       "\n",
       "<style>\n",
       "\n",
       "        .variablerow {\n",
       "            border: 1px solid #e1e1e8;\n",
       "            border-top: hidden;\n",
       "            padding-top: 2em;\n",
       "            padding-bottom: 2em;\n",
       "            padding-left: 1em;\n",
       "            padding-right: 1em;\n",
       "        }\n",
       "\n",
       "        .headerrow {\n",
       "            border: 1px solid #e1e1e8;\n",
       "            background-color: #f5f5f5;\n",
       "            padding: 2em;\n",
       "        }\n",
       "        .namecol {\n",
       "            margin-top: -1em;\n",
       "            overflow-x: auto;\n",
       "        }\n",
       "\n",
       "        .dl-horizontal dt {\n",
       "            text-align: left;\n",
       "            padding-right: 1em;\n",
       "            white-space: normal;\n",
       "        }\n",
       "\n",
       "        .dl-horizontal dd {\n",
       "            margin-left: 0;\n",
       "        }\n",
       "\n",
       "        .ignore {\n",
       "            opacity: 0.4;\n",
       "        }\n",
       "\n",
       "        .container.pandas-profiling {\n",
       "            max-width:975px;\n",
       "        }\n",
       "\n",
       "        .col-md-12 {\n",
       "            padding-left: 2em;\n",
       "        }\n",
       "\n",
       "        .indent {\n",
       "            margin-left: 1em;\n",
       "        }\n",
       "\n",
       "        .center-img {\n",
       "            margin-left: auto !important;\n",
       "            margin-right: auto !important;\n",
       "            display: block;\n",
       "        }\n",
       "\n",
       "        /* Table example_values */\n",
       "            table.example_values {\n",
       "                border: 0;\n",
       "            }\n",
       "\n",
       "            .example_values th {\n",
       "                border: 0;\n",
       "                padding: 0 ;\n",
       "                color: #555;\n",
       "                font-weight: 600;\n",
       "            }\n",
       "\n",
       "            .example_values tr, .example_values td{\n",
       "                border: 0;\n",
       "                padding: 0;\n",
       "                color: #555;\n",
       "            }\n",
       "\n",
       "        /* STATS */\n",
       "            table.stats {\n",
       "                border: 0;\n",
       "            }\n",
       "\n",
       "            .stats th {\n",
       "                border: 0;\n",
       "                padding: 0 2em 0 0;\n",
       "                color: #555;\n",
       "                font-weight: 600;\n",
       "            }\n",
       "\n",
       "            .stats tr {\n",
       "                border: 0;\n",
       "            }\n",
       "\n",
       "            .stats td{\n",
       "                color: #555;\n",
       "                padding: 1px;\n",
       "                border: 0;\n",
       "            }\n",
       "\n",
       "\n",
       "        /* Sample table */\n",
       "            table.sample {\n",
       "                border: 0;\n",
       "                margin-bottom: 2em;\n",
       "                margin-left:1em;\n",
       "            }\n",
       "            .sample tr {\n",
       "                border:0;\n",
       "            }\n",
       "            .sample td, .sample th{\n",
       "                padding: 0.5em;\n",
       "                white-space: nowrap;\n",
       "                border: none;\n",
       "\n",
       "            }\n",
       "\n",
       "            .sample thead {\n",
       "                border-top: 0;\n",
       "                border-bottom: 2px solid #ddd;\n",
       "            }\n",
       "\n",
       "            .sample td {\n",
       "                width:100%;\n",
       "            }\n",
       "\n",
       "\n",
       "        /* There is no good solution available to make the divs equal height and then center ... */\n",
       "            .histogram {\n",
       "                margin-top: 3em;\n",
       "            }\n",
       "        /* Freq table */\n",
       "\n",
       "            table.freq {\n",
       "                margin-bottom: 2em;\n",
       "                border: 0;\n",
       "            }\n",
       "            table.freq th, table.freq tr, table.freq td {\n",
       "                border: 0;\n",
       "                padding: 0;\n",
       "            }\n",
       "\n",
       "            .freq thead {\n",
       "                font-weight: 600;\n",
       "                white-space: nowrap;\n",
       "                overflow: hidden;\n",
       "                text-overflow: ellipsis;\n",
       "\n",
       "            }\n",
       "\n",
       "            td.fillremaining{\n",
       "                width:auto;\n",
       "                max-width: none;\n",
       "            }\n",
       "\n",
       "            td.number, th.number {\n",
       "                text-align:right ;\n",
       "            }\n",
       "\n",
       "        /* Freq mini */\n",
       "            .freq.mini td{\n",
       "                width: 50%;\n",
       "                padding: 1px;\n",
       "                font-size: 12px;\n",
       "\n",
       "            }\n",
       "            table.freq.mini {\n",
       "                 width:100%;\n",
       "            }\n",
       "            .freq.mini th {\n",
       "                overflow: hidden;\n",
       "                text-overflow: ellipsis;\n",
       "                white-space: nowrap;\n",
       "                max-width: 5em;\n",
       "                font-weight: 400;\n",
       "                text-align:right;\n",
       "                padding-right: 0.5em;\n",
       "            }\n",
       "\n",
       "            .missing {\n",
       "                color: #a94442;\n",
       "            }\n",
       "            .alert, .alert > th, .alert > td {\n",
       "                color: #a94442;\n",
       "            }\n",
       "\n",
       "\n",
       "        /* Bars in tables */\n",
       "            .freq .bar{\n",
       "                float: left;\n",
       "                width: 0;\n",
       "                height: 100%;\n",
       "                line-height: 20px;\n",
       "                color: #fff;\n",
       "                text-align: center;\n",
       "                background-color: #337ab7;\n",
       "                border-radius: 3px;\n",
       "                margin-right: 4px;\n",
       "            }\n",
       "            .other .bar {\n",
       "                background-color: #999;\n",
       "            }\n",
       "            .missing .bar{\n",
       "                background-color: #a94442;\n",
       "            }\n",
       "            .tooltip-inner {\n",
       "                width: 100%;\n",
       "                white-space: nowrap;\n",
       "                text-align:left;\n",
       "            }\n",
       "\n",
       "            .extrapadding{\n",
       "                padding: 2em;\n",
       "            }\n",
       "\n",
       "            .pp-anchor{\n",
       "\n",
       "            }\n",
       "\n",
       "</style>\n",
       "\n",
       "<div class=\"container pandas-profiling\">\n",
       "    <div class=\"row headerrow highlight\">\n",
       "        <h1>Overview</h1>\n",
       "    </div>\n",
       "    <div class=\"row variablerow\">\n",
       "    <div class=\"col-md-6 namecol\">\n",
       "        <p class=\"h4\">Dataset info</p>\n",
       "        <table class=\"stats\" style=\"margin-left: 1em;\">\n",
       "            <tbody>\n",
       "            <tr>\n",
       "                <th>Number of variables</th>\n",
       "                <td>7 </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Number of observations</th>\n",
       "                <td>481 </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Total Missing (%)</th>\n",
       "                <td>0.0% </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Total size in memory</th>\n",
       "                <td>26.4 KiB </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Average record size in memory</th>\n",
       "                <td>56.2 B </td>\n",
       "            </tr>\n",
       "            </tbody>\n",
       "        </table>\n",
       "    </div>\n",
       "    <div class=\"col-md-6 namecol\">\n",
       "        <p class=\"h4\">Variables types</p>\n",
       "        <table class=\"stats\" style=\"margin-left: 1em;\">\n",
       "            <tbody>\n",
       "            <tr>\n",
       "                <th>Numeric</th>\n",
       "                <td>3 </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Categorical</th>\n",
       "                <td>1 </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Boolean</th>\n",
       "                <td>0 </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Date</th>\n",
       "                <td>0 </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Text (Unique)</th>\n",
       "                <td>0 </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Rejected</th>\n",
       "                <td>2 </td>\n",
       "            </tr>\n",
       "            <tr>\n",
       "                <th>Unsupported</th>\n",
       "                <td>1 </td>\n",
       "            </tr>\n",
       "            </tbody>\n",
       "        </table>\n",
       "    </div>\n",
       "    <div class=\"col-md-12\" style=\"padding-left: 1em;\">\n",
       "        \n",
       "        <p class=\"h4\">Warnings</p>\n",
       "        <ul class=\"list-unstyled\"><li><a href=\"#pp_var_fs\"><code>fs</code></a> has constant value 16000 <span class=\"label label-primary\">Rejected</span></li><li><a href=\"#pp_var_ind_numb\"><code>ind_numb</code></a> is highly correlated with <a href=\"#pp_var_index\"><code>index</code></a> (ρ = 0.95218) <span class=\"label label-primary\">Rejected</span></li><li><a href=\"#pp_var_x\"><code>x</code></a> is an unsupported type, check if it needs cleaning or further analysis <span class=\"label label-warning\">Warning</span></li><li><a href=\"#pp_var_y\"><code>y</code></a> has 123 / 25.6% zeros <span class=\"label label-info\">Zeros</span></li> </ul>\n",
       "    </div>\n",
       "</div>\n",
       "    <div class=\"row headerrow highlight\">\n",
       "        <h1>Variables</h1>\n",
       "    </div>\n",
       "    <div class=\"row variablerow ignore\">\n",
       "    <div class=\"col-md-3 namecol\">\n",
       "        <p class=\"h4 pp-anchor\" id=\"pp_var_fs\"><s>fs</s><br/>\n",
       "            <small>Constant</small>\n",
       "        </p>\n",
       "    </div><div class=\"col-md-3\">\n",
       "    <p><em>This variable is constant and should be ignored for analysis</em></p>\n",
       "</div>\n",
       "<div class=\"col-md-6\">\n",
       "    <table class=\"stats \">\n",
       "        <tr>\n",
       "            <th>Constant value</th>\n",
       "            <td>16000</td>\n",
       "        </tr>\n",
       "    </table>\n",
       "</div>\n",
       "</div><div class=\"row variablerow ignore\">\n",
       "    <div class=\"col-md-3 namecol\">\n",
       "        <p class=\"h4 pp-anchor\" id=\"pp_var_ind_numb\"><s>ind_numb</s><br/>\n",
       "            <small>Highly correlated</small>\n",
       "        </p>\n",
       "    </div><div class=\"col-md-3\">\n",
       "    <p><em>This variable is highly correlated with <a href=\"#pp_var_index\"><code>index</code></a> and should be ignored for analysis</em></p>\n",
       "</div>\n",
       "<div class=\"col-md-6\">\n",
       "    <table class=\"stats \">\n",
       "        <tr>\n",
       "            <th>Correlation</th>\n",
       "            <td>0.95218</td>\n",
       "        </tr>\n",
       "    </table>\n",
       "</div>\n",
       "</div><div class=\"row variablerow\">\n",
       "    <div class=\"col-md-3 namecol\">\n",
       "        <p class=\"h4 pp-anchor\" id=\"pp_var_index\">index<br/>\n",
       "            <small>Numeric</small>\n",
       "        </p>\n",
       "    </div><div class=\"col-md-6\">\n",
       "    <div class=\"row\">\n",
       "        <div class=\"col-sm-6\">\n",
       "            <table class=\"stats \">\n",
       "                <tr>\n",
       "                    <th>Distinct count</th>\n",
       "                    <td>481</td>\n",
       "                </tr>\n",
       "                <tr>\n",
       "                    <th>Unique (%)</th>\n",
       "                    <td>100.0%</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Missing (%)</th>\n",
       "                    <td>0.0%</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Missing (n)</th>\n",
       "                    <td>0</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Infinite (%)</th>\n",
       "                    <td>0.0%</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Infinite (n)</th>\n",
       "                    <td>0</td>\n",
       "                </tr>\n",
       "            </table>\n",
       "\n",
       "        </div>\n",
       "        <div class=\"col-sm-6\">\n",
       "            <table class=\"stats \">\n",
       "\n",
       "                <tr>\n",
       "                    <th>Mean</th>\n",
       "                    <td>269.12</td>\n",
       "                </tr>\n",
       "                <tr>\n",
       "                    <th>Minimum</th>\n",
       "                    <td>0</td>\n",
       "                </tr>\n",
       "                <tr>\n",
       "                    <th>Maximum</th>\n",
       "                    <td>534</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Zeros (%)</th>\n",
       "                    <td>0.2%</td>\n",
       "                </tr>\n",
       "            </table>\n",
       "        </div>\n",
       "    </div>\n",
       "</div>\n",
       "<div class=\"col-md-3 collapse in\" id=\"minihistogram3896701726749123897\">\n",
       "    <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAMgAAABLCAYAAAA1fMjoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD%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%3D%3D\">\n",
       "\n",
       "</div>\n",
       "<div class=\"col-md-12 text-right\">\n",
       "    <a role=\"button\" data-toggle=\"collapse\" data-target=\"#descriptives3896701726749123897,#minihistogram3896701726749123897\"\n",
       "       aria-expanded=\"false\" aria-controls=\"collapseExample\">\n",
       "        Toggle details\n",
       "    </a>\n",
       "</div>\n",
       "<div class=\"row collapse col-md-12\" id=\"descriptives3896701726749123897\">\n",
       "    <ul class=\"nav nav-tabs\" role=\"tablist\">\n",
       "        <li role=\"presentation\" class=\"active\"><a href=\"#quantiles3896701726749123897\"\n",
       "                                                  aria-controls=\"quantiles3896701726749123897\" role=\"tab\"\n",
       "                                                  data-toggle=\"tab\">Statistics</a></li>\n",
       "        <li role=\"presentation\"><a href=\"#histogram3896701726749123897\" aria-controls=\"histogram3896701726749123897\"\n",
       "                                   role=\"tab\" data-toggle=\"tab\">Histogram</a></li>\n",
       "        <li role=\"presentation\"><a href=\"#common3896701726749123897\" aria-controls=\"common3896701726749123897\"\n",
       "                                   role=\"tab\" data-toggle=\"tab\">Common Values</a></li>\n",
       "        <li role=\"presentation\"><a href=\"#extreme3896701726749123897\" aria-controls=\"extreme3896701726749123897\"\n",
       "                                   role=\"tab\" data-toggle=\"tab\">Extreme Values</a></li>\n",
       "\n",
       "    </ul>\n",
       "\n",
       "    <div class=\"tab-content\">\n",
       "        <div role=\"tabpanel\" class=\"tab-pane active row\" id=\"quantiles3896701726749123897\">\n",
       "            <div class=\"col-md-4 col-md-offset-1\">\n",
       "                <p class=\"h4\">Quantile statistics</p>\n",
       "                <table class=\"stats indent\">\n",
       "                    <tr>\n",
       "                        <th>Minimum</th>\n",
       "                        <td>0</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>5-th percentile</th>\n",
       "                        <td>27</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Q1</th>\n",
       "                        <td>134</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Median</th>\n",
       "                        <td>272</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Q3</th>\n",
       "                        <td>407</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>95-th percentile</th>\n",
       "                        <td>507</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Maximum</th>\n",
       "                        <td>534</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Range</th>\n",
       "                        <td>534</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Interquartile range</th>\n",
       "                        <td>273</td>\n",
       "                    </tr>\n",
       "                </table>\n",
       "            </div>\n",
       "            <div class=\"col-md-4 col-md-offset-2\">\n",
       "                <p class=\"h4\">Descriptive statistics</p>\n",
       "                <table class=\"stats indent\">\n",
       "                    <tr>\n",
       "                        <th>Standard deviation</th>\n",
       "                        <td>155.95</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Coef of variation</th>\n",
       "                        <td>0.57949</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Kurtosis</th>\n",
       "                        <td>-1.2363</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Mean</th>\n",
       "                        <td>269.12</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>MAD</th>\n",
       "                        <td>135.82</td>\n",
       "                    </tr>\n",
       "                    <tr class=\"\">\n",
       "                        <th>Skewness</th>\n",
       "                        <td>-0.023081</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Sum</th>\n",
       "                        <td>129448</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Variance</th>\n",
       "                        <td>24321</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Memory size</th>\n",
       "                        <td>3.8 KiB</td>\n",
       "                    </tr>\n",
       "                </table>\n",
       "            </div>\n",
       "        </div>\n",
       "        <div role=\"tabpanel\" class=\"tab-pane col-md-8 col-md-offset-2\" id=\"histogram3896701726749123897\">\n",
       "            <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAlgAAAGQCAYAAAByNR6YAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD%2BnaQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3X9U1vX9//HHBVfQQAGx0KYVLgfGD8mfFNVILE2bzhRFq5NOK1PKydGm6fykZcqOcvo26Wxap23l2byGZ6WSv06ZZhv7YbkGqCx/tDXCIL2uIQRO4P39w%2BPVLtHUeOmbN9f9do6n%2BXpfvn3yOhde972vH7gsy7IEAAAAY0LsHgAAAKCjIbAAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMc9s9QLCoqTlh/JwhIS7Fxkbq%2BPF6tbRYxs8fLNjHtmMPzWAfzWAf264j7eG113a25e/lCpaDhYS45HK5FBLisnsUR2Mf2449NIN9NIN9bDv2sO0ILAAAAMMILAAAAMMILAAAAMMILAAAAMMILAAAAMMILAAAAMMILAAAAMMILAAAAMMILAAAAMMILAAAAMMILAAAAMMILAAAAMMILAAAAMPcdg8AAAheI/7fH%2BweoUPbMvt2u0cIWlzBAgAAMCwor2AlJibqqquuksvl8q9NmDBBixYtUklJiQoKCnT48GFdd911mj59ukaPHm3jtAAAwGmCMrAkaevWrerZs2fAWnV1tWbOnKmFCxdq1KhR%2BuCDDzRjxgz16tVLqampNk0KAACchqcI/8emTZsUHx%2Bv7OxshYeHKyMjQ1lZWSoqKrJ7NAAA4CBBewWroKBAe/fuVV1dnUaMGKH58%2BervLxcSUlJAbdLSkrSli1bLunc1dXVqqmpCVhzuyMUFxfX5rn/V2hoSMB/8c2wj23HHprBPsI0t/ub3Ze4L7ZdUAbWLbfcooyMDP30pz/Vp59%2BqtmzZ2vJkiXy%2BXzq1q1bwG1jYmLk9Xov6fwej0eFhYUBa7m5uZo1a1abZz%2BXqKhvXZbzBhv2se3YQzPYR5jSpUtkm/4898VvLigDy%2BPx%2BP/3TTfdpLlz52rGjBkaMGCAkfPn5OQoKysrYM3tjpDXW2/k/GeEhoYoKupbqq1tUHNzi9FzBxP2se3YQzPYR5j2TR93OtJ9sa2R%2BU0FZWCdrWfPnmpublZISIh8Pl/AMa/Xq9jY2Es6X1xcXKunA2tqTqip6fLcSZubWy7buYMJ%2B9h27KEZ7CNMaev9iPviNxd0T67u27dP%2Bfn5AWuHDh1SWFiYMjMzVVZWFnCsrKxMaWlpV3JEAADgcEEXWF27dpXH49GaNWv03//%2BV0eOHNGLL76onJwc/eAHP1BlZaWKiop08uRJ7dq1S7t27dKECRPsHhsAADhI0AVWt27dtGbNGu3YsUPp6emaOHGi7rzzTj311FPq2rWrVq9erbVr12rAgAFatmyZVqxYoT59%2Btg9NgAAcBCXZVmW3UMEg5qaE8bP6XaH6J6Vu42fF6fxM7wuntsdoi5dIuX11vN6jTYIxn3kZxHijMv1b%2B6113a%2BLOe9kKC7ggUAAHC5EVgAAACGEVgAAACGEVgAAACGEVgAAACG8UnuQAfhpHdjOe0dmk7aW8l5%2Bwt0RFzBAgAAMIzAAgAAMIzAAgAAMIzAAgAAMIzAAgAAMIzAAgAAMIzAAgAAMIzAAgAAMIzAAgAAMIzAAgAAMIzAAgAAMIyfRQich9N%2B/hxwBvddwH5cwQIAADCMwAIAADCMwAIAADCMwAIAADCMwAIAADCMwAIAADCMwAIAADCMwAIAADCMwAIAADCMwAIAADCMwAIAADCMn0UI4IrjZ%2BUB6Oi4ggUAAGAYgQUAAGAYgQUAAGAYgQUAAGAYgQUAAGAYgQUAAGAYgQUAAGAYgQUAAGAYgQUAAGAYgQUAAGAYgQUAAGAYgQUAAGAYgQUAAGAYgQUAAGAYgQUAAGAYgQUAAGAYgQUAAGAYgQUAAGBY0AfWsmXLlJiY6P99SUmJsrOz1b9/f913333auHGjjdMBAAAncts9gJ3279%2BvDRs2%2BH9fXV2tmTNnauHChRo1apQ%2B%2BOADzZgxQ7169VJqaqqNkwIAACcJ2itYLS0teuaZZzRlyhT/2qZNmxQfH6/s7GyFh4crIyNDWVlZKioqsm9QAADgOEEbWOvWrVN4eLhGjRrlXysvL1dSUlLA7ZKSklRWVnalxwMAAA4WlE8RfvHFF1q1apVef/31gHWfz6du3boFrMXExMjr9V7S%2Baurq1VTUxOw5nZHKC4u7psNfB6hoUHbxwCADsbt7liPaUEZWMuXL9fYsWPVu3dv/fvf/zZ%2Bfo/Ho8LCwoC13NxczZo1y/jfBQBAR9ClS6TdIxgVdIFVUlKivXv3qri4uNWxLl26yOfzBax5vV7FxsZe0t%2BRk5OjrKysgDW3O0Jeb/2lD/w1uIIFAOgoTD9GnmFXuAVdYG3cuFHHjh3TkCFDJEmWZUmS0tPTNXXq1FbhVVZWprS0tEv6O%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%2B7Rzqum5oTxc7rdIbpn5W7j5wUA4ErbMvv2y3Lea6/tfFnOeyGODKwzLMvS5s2btXjxYtXV1SkjI0M/%2BtGP1LdvX7tHa4XAAgDg/DpaYDnqKcIzTp06pc2bN%2BvRRx/VvHnz1K1bNz399NO6%2BeabNWXKFG3atMnuEQEAQBBz2z3ApTh06JDWr1%2BvN998U/X19Ro%2BfLh%2B/etfa8CAAf7bDBo0SIsXL9aoUaNsnBQAAAQzRwXWfffdp169emn69OkaM2aMYmJiWt0mMzNTx48ft2E6AACA0xwVWK%2B99poGDx58wdt99NFHV2AaAACAc3PUa7ASExP1%2BOOP6%2B233/av/epXv9Kjjz4qn89n42QAAABfcVRgLV%2B%2BXCdOnFDv3r39a3fddZdaWlqUn59/Sec6cOCAJk%2BerAEDBigjI0OzZ89WTU2NJKmkpETZ2dnq37%2B/7rvvPm3cuNHo1wEAADo2RwXW%2B%2B%2B/r8LCQsXHx/vX4uPjtXLlSu3effEfV/Df//5XU6dO1eDBg1VSUqLi4mIdO3ZMixcvVnV1tWbOnKmJEyeqpKRECxcu1KJFi1RaWnoZviIAANAROSqwGhsbFR4e3mo9JCREDQ0NF32ehoYG5eXlafr06QoLC1NsbKzuueceffzxx9q0aZPi4%2BOVnZ2t8PBwZWRkKCsrS0VFRSa/FAAA0IE5KrAGDRqk/Px8/ec///Gvff7551qyZEnARzVcSHR0tMaPHy%2B3%2B/Rr/A8fPqw33nhDI0aMUHl5uZKSkgJun5SUpLKyMjNfBAAA6PAc9S7CBQsWaOrUqbrtttvUqVMntbS0qL6%2BXtdff71ef/31Sz5fZWWlhg8frqamJk2YMEGzZs3So48%2Bqm7dugXcLiYmRl6v96LPW11d7X891xlud4Ti4uIuecavExrqqD4GAOC83O6O9ZjmqMC6/vrr9dZbb%2Bm9997Tv/71L4WEhKhXr1664447FBoaesnn69Gjh0pLS/XPf/5T//d//6cf//jHRub0eDwqLCwMWMvNzdWsWbOMnB8AgI6mS5dIu0cwylGBJUlhYWG6%2B%2B67jZ3P5XIpPj5eeXl5mjhxojIzM1t95IPX61VsbOxFnzMnJ0dZWVkBa253hLzeeiMzn8EVLABAR2H6MfIMu8LNUYH16aefqqCgQB9//LEaGxtbHX/nnXcu6jwlJSVavHixtmzZopCQ05Fy5r99%2B/bVtm3bAm5fVlamtLS0i54zLi6u1dOBNTUn1NTUctHnAAAgmHS0x0hHBdaCBQtUXV2tO%2B64QxEREd/4PCkpKaqrq9OKFSs0a9YsNTQ0aNWqVRo4cKAmTZqkV199VUVFRRo9erT%2B9Kc/adeuXfJ4PAa/EgAA0JE5KrDKysr0zjvvXNLTdefSuXNnvfrqq1q6dKluvfVWRURE6NZbb9Xzzz%2Bvrl27avXq1Vq6dKmWLFmiHj16aMWKFerTp4%2BhrwIAAHR0jgqsrl27tunK1f9KTEw87zsPBw0apA0bNhj5ewAAQPBx1Kukp0%2BfrsLCQlmWZfcoAAAA5%2BWoK1jvvfeePvzwQ/3%2B979Xz549/S9MP2PdunU2TQYAAPAVRwVWp06d9L3vfc/uMQAAAL6WowJr%2BfLldo8AAABwQY56DZZ0%2BucGrlq1Sk8//bR/be/evTZOBAAAEMhRgVVSUqLRo0dr%2B/btKi4ulnT6w0cffvjhi/6QUQAAgMvNUYH1wgsv6KmnntKmTZvkcrkknf75hPn5%2BXrppZdsng4AAOA0RwXWP/7xD02aNEmS/IElSffee68OHTpk11gAAAABHBVYnTt3PufPIKyurlZYWJgNEwEAALTmqMDq37%2B/li1bprq6Ov/akSNHNG/ePN122202TgYAAPAVR31Mw9NPP63JkycrPT1dzc3N6t%2B/vxoaGvTd735X%2Bfn5do8HAAAgyWGB1b17dxUXF2vXrl06cuSIrr76avXq1Uu33357wGuyAAAA7OSowJKkq666SnfffbfdYwAAAJyXowIrKyvra69U8VlYAACgPXBUYI0cOTIgsJqbm3XkyBGVlpZq8uTJNk4GAADwFUcF1ty5c8%2B5vm3bNv35z3%2B%2BwtMAAACcm6M%2BpuF87r77br311lt2jwEAACCpgwTWvn37ZFmW3WMAAABIcthThBMnTmy11tDQoEOHDmnYsGE2TAQAANCaowIrPj6%2B1bsIw8PDlZ2drfHjx9s0FQAAQCBHBRaf1g4AAJzAUYH15ptvXvRtx4wZcxknAQAAOD9HBdbChQvV0tLS6gXtLpcrYM3lchFYAADANo4KrFdeeUWvvvqqHn/8cSUmJsqyLFVUVOjll1/WQw89pPT0dLtHBAAAcFZg5efna82aNerWrZt/beDAgbr%2B%2Bus1bdo0FRcX2zgdAADAaY76HKxPPvlE0dHRrdajoqJUWVlpw0QAAACtOSqwevToofz8fHm9Xv9abW2tCgoKdMMNN9g4GQAAwFcc9RThggULNGfOHHk8HkVGRiokJER1dXW6%2Buqr9dJLL9k9HgAAgCSHBdYdd9yhnTt3ateuXTp69Kgsy1K3bt105513qnPnznaPBwAAIMlhgSVJ3/rWtzR06FAdPXpU119/vd3jAAAAtOKo12A1NjZq3rx56tevn0aMGCHp9GuwHnnkEdXW1to8HQAAwGmOCqwVK1Zo//79WrlypUJCvhq9ublZK1eutHEyAACArzgqsLZt26af/exnuvfee/0/9DkqKkrLly/X9u3bbZ4OAADgNEcFVn19veLj41utx8bG6ssvv7zyAwEAAJyDowLrhhtu0J///GdJCvjZg1u3btW3v/1tu8YCAAAI4Kh3ET7wwAN68sknNW7cOLW0tOiXv/ylysrKtG3bNi1cuNDu8QAAACQ5LLBycnLkdru1du1ahYaG6he/%2BIV69eqllStX6t5777V7PAAAAEkOC6zjx49r3LhxGjdunN2jAAAAnJejXoM1dOjQgNdeAQAAtEeOCqz09HRt2bLF7jEAAAC%2BlqOeIrzuuuv0/PPPa82aNbrhhht01VVXBRwvKCiwaTIAAICvOCqwDh48qO985zuSJK/Xa/M0AAAA5%2BaIwMrLy9MLL7yg119/3b/20ksvKTc318apAAAAzs0Rr8HasWNHq7U1a9bYMAkAAMCFOSKwzvXOQd5NCAAA2itHBNaZH%2Bx8oTUAAID2wBGBBQAA4CQEFgAAgGGOeBfhqVOnNGfOnAuu8TlYAACgPXBEYA0YMEDV1dUXXLsUlZWVWrZsmfbs2aPQ0FB973vf04IFCxQVFaX9%2B/fr%2Beef1/79%2B9W1a1dNnDhRU6dObeuXAQAAgoQjAut/P//KlMcff1wpKSnasWOHTpw4odzcXP30pz/VokWLNH36dE2YMEFr1qzRkSNHNHXqVPXs2VPDhg0zPgcAAOh4gvI1WLW1tUpJSdGcOXMUGRmp7t276/7779eePXu0c%2BdOnTp1SjNmzFBERISSk5M1fvx4eTweu8cGAAAO4YgrWKZFRUVp%2BfLlAWtVVVWKi4tTeXm5EhMTFRoa6j%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%2BWc577bWdL8t5LyQoX4M1cOBAVVRUfO1tfvvb316haQAAQEcTlE8RAgAAXE4EFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEEFgAAgGEE1jns3r1bGRkZysvLa3Vs8%2BbNGjVqlPr166exY8fq/ffft2FCAADQnrntHqC9efnll7V%2B/XrdeOONrY7t379f8%2BbNU2FhoW699VZt27ZNTzzxhLZu3aru3bvbMC0AAGiPuIJ1lvDw8PMGVlFRkTIzM5WZmanw8HCNHj1aCQkJ2rhxow2TAgCA9oorWGd5%2BOGHz3usvLxcmZmZAWtJSUkqLS0NWKuurlZNTU3Amtsdobi4OHODSgoNpY8BAB2D292xHtMIrEvg8/kUHR0dsBYdHa2DBw8GrHk8HhUWFgas5ebmatasWZd9RgAAnKhLl0i7RzCKwLpElmVd8DY5OTnKysoKWHO7I%2BT11hudhStYAICOwvRj5Bl2hRuBdQm6dOkin88XsObz%2BRQbGxuwFhcX1%2BrpwJqaE2pqarnsMwIA4EQd7TGSSyCXICUlRWVlZQFrpaWlSktLs2kiAADQHhFYl2DChAn64x//qJ07d%2BrkyZNav369PvnkE40ePdru0QAAQDvCU4RnSU1NlSQ1NTVJkt5%2B%2B21Jp69UJSQkaOXKlVq%2BfLkqKyvVu3dvrV69Wtdee61t8wIAgPaHwDrL2R%2B5cLZhw4Zp2LBhV2gaAADgRDxFCAAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBBQAAYBiBdQ6VlZV67LHHlJ6eriFDhmjFihVqaWmxeywAAOAQbrsHaI%2BefPJJJScn6%2B2339axY8c0ffp0XXPNNfrhD39o92gAAMABuIJ1ltLSUh04cEBz585V586dFR8frylTpsjj8dg9GgAAcAiuYJ2lvLxcPXr0UHR0tH8tOTlZR44cUV1dnTp16nTBc1RXV6umpiZgze2OUFxcnNFZQ0PpYwBAx%2BB2d6zHNALrLD6fT1FRUQFrZ2LL6/VeVGB5PB4VFhYGrD3xxBN68sknzQ2q0yE3ufvHysnJMR5vwaS6uloej4d9bAP20Az20Qz2se3Yw7brWLloiGVZbfrzOTk5%2Bv3vfx/wKycnx9B0X6mpqVFhYWGrq2W4NOxj27GHZrCPZrCPbcceth1XsM4SGxsrn88XsObz%2BeRyuRQbG3tR54iLi6P4AQAIYlzBOktKSoqqqqp0/Phx/1ppaal69%2B6tyMhIGycDAABOQWCdJSkpSampqSooKFBdXZ0OHTqkX/7yl5o0aZLdowEAAIcIXbx48WK7h2hv7rzzThUXF%2Bu5557TW2%2B9pezsbE2bNk0ul8vu0VqJjIzU4MGDubrWRuxj27GHZrCPZrCPbcceto3LausrugEAABCApwgBAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AAAAAMI7AcqLKyUo899pjS09M1ZMgQrVixQi0tLXaP1S7t3r1bGRkZysvLa3Vs8%2BbNGjVqlPr166exY8fq/fff9x9raWnRCy%2B8oKFDh2rQoEGaNm2aPv300ys5ertRWVmp3NxcpaenKyMjQ/Pnz1dtba0kadeuXRo%2BfLjS0tJ08803KzExUV6v1%2BaJ26cDBw5o8uTJGjBggDIyMjR79mzV1NRIOn1fTE5OVnJysgYNGqShQ4eqsbHR5onbt2XLlikxMdH/%2B5KSEmVnZ6t///667777tHHjxoDbv/baaxo%2BfLj69%2B%2BvSZMmqays7EqP3G4kJiYqJSVFqamp/l/PPfecJPbRKAuOc//991s/%2BclPrNraWuvIkSPWsGHDrFdffdXusdqdNWvWWMOGDbMmTpxozZ49O%2BDYvn37rJSUFGvnzp1WY2OjtWHDBistLc2qqqqyLMuyXnvtNWvIkCHWwYMHrRMnTljPPvusNWrUKKulpcWOL8VW3//%2B96358%2BdbdXV1VlVVlTV27FhrwYIF1ueff2716dPHSk9Ptz777DPrrbfeshISEqxHHnnE7pHbnZMnT1q33XabVVhYaJ08edI6duyY9dBDD1kzZ8609u3bZ/Xp08dKTEy0qqqq/PfF%2BfPn2z12u7Vv3z5r8ODBVkJCgmVZlvX5559bt9xyi1VUVGQ1NjZaf/jDH6y%2Bfftaf//73y3Lsqx33nnHGjhwoPW3v/3NamhosFavXm3dfvvtVn19vZ1fhm0SEhKsTz/9tNU6%2B2gWV7AcprS0VAcOHNDcuXPVuXNnxcfHa8qUKfJ4PHaP1u6Eh4dr/fr1uvHGG1sdKyoqUmZmpjIzMxUeHq7Ro0crISHB///WPB6PpkyZoptuukmdOnVSXl6eDh06pI8%2B%2BuhKfxm2qq2tVUpKiubMmaPIyEh1795d999/v/bs2aNNmzbJsiw9//zzuu666zRy5EiNHDlSu3fv1ueff2736O1KQ0OD8vLyNH36dIWFhSk2Nlb33HOPPv74Y61bt04tLS1688031b1794D74qlTp%2Bwevd1paWnRM888oylTpvjXNm3apPj4eGVnZys8PFwZGRnKyspSUVGRpNPfz2PHjlVaWpquvvpqPfLII5Kkd999144vod1iH80isBymvLxcPXr0UHR0tH8tOTlZR44cUV1dnY2TtT8PP/ywOnfufM5j5eXlSkpKClhLSkpSaWmpGhsbdfDgwYDjnTp10o033qjS0tLLOnN7ExUVpeXLl%2Buaa67xr1VVVSkuLk7l5eWyLEvJycn%2BY0lJSXK5XCovL7dj3HYrOjpa48ePl9vtliQdPnxYb7zxhkaMGKEDBw7I5XIFPN2VlJSkpqYmHT582K6R261169YpPDxco0aN8q%2Bd7/v5zNNXZx8PCQnRzTffHHTfz/%2BroKBAd911lwYOHKhFixapvr6efTSMwHIYn8%2BnqKiogLUzscUGf64lAAAFJUlEQVRrXy6ez%2BcLiFTp9D56vV795z//kWVZ5z0ezEpLS7V27VrNmDFDPp9PkgLujzExMbrmmmuCfp/Op7KyUikpKRo5cqRSU1M1a9Ys%2BXw%2BxcTEyOVy%2BW/H9/S5ffHFF1q1apWeeeaZgPVz/bsYExPj37%2Bv%2B34PRrfccosyMjK0fft2eTwe/e1vf9OSJUvYR8MILAeyLMvuETqEC%2B0j%2Bxzogw8%2B0LRp0zRnzhxlZGT419mni9ejRw%2BVlpZq69at%2BuSTT/TjH/9YEnt4sZYvX66xY8eqd%2B/el/xn2eOveDwejR8/XmFhYbrppps0d%2B5cFRcXX9RT0uzjxSOwHCY2NtZ/5eAMn88nl8ul2NhYm6Zyni5dupxzH2NjYxUTE6OQkJBzHu/ateuVHLPd2LFjhx577DEtWLBADz/8sKTTeygpYJ%2B8Xq%2Bqq6uDdp8uhsvlUnx8vPLy8lRcXCy32y2fz6fm5mb/bc7sKfv4lZKSEu3du1e5ubmtjp3r%2B9nr9fr/Tfy673dIPXv2VHNz8zn/3WMfvzkCy2FSUlJUVVWl48eP%2B9dKS0vVu3dvRUZG2jiZs6SkpLR6e3FpaanS0tIUHh6u7373uwGvI6qtrdW//vUv9e3b90qParsPP/xQ8%2BbN04svvqgxY8b415OTk1u93qqsrEwhISFKSUmxY9R2q6SkRMOHDw/4OJWQkNP//Pbt21cul0sHDhzwHzvzmpZevXpd2UHbsY0bN%2BrYsWMaMmSI0tPTNXbsWElSenq6EhISWn0/l5WVKS0tTdLp7/f/vZ82Nzdr3759/uPBZN%2B%2BfcrPzw9YO3TokMLCwpSZmck%2BmmTX2xfxzY0fP95asGCBdeLECevgwYNWVlaWtXbtWrvHarfmzZvX6mMaKioqrNTUVOvdd9%2B1GhsbraKiIqtfv35WdXW1ZVmW9Zvf/Ma66667/B/TsGjRImvcuHF2jG%2BrU6dOWSNGjLDWrVvX6tgXX3xhJScnW4MHD7Y%2B%2B%2Bwza8OGDVZqaqo1c%2BZMGyZt32pra62MjAwrPz/f%2BvLLL61jx45Z06ZNsx544AGroqLCSkhIsBISEqyqqipr1apVVnJysrVw4UK7x25XfD6fVVVV5f%2B1d%2B9e/55VVlZa/fr1s373u99ZjY2N1s6dO62%2Bffta%2B/fvtyzLsnbt2mUNGDDA2rt3r/Xll19aq1atsjIzM62Ghgabv6or7%2BjRo9Ytt9xirV692jp58qR1%2BPBha%2BTIkdZzzz1nffHFF%2ByjQS7L4glVpzl69KgWLVqkv/zlL%2BrUqZMmTpyoJ554IuBFspBSU1MlSU1NTZLkfwfXmasD27dvV0FBgSorK9W7d28tXLhQgwYNknT6dQarVq3SunXrVF9fr/T0dD377LPq3r27DV%2BJffbs2aMHH3xQYWFhrY5t3bpVn332mRYvXqyDBw/K7XbL7Xbr/fffP%2B%2B7N4NZRUWFli5dqr///e%2BKiIjQrbfeqvnz56tbt2564403NH/%2BfLndbkVEROjmm2/WK6%2B8cs59x2n//ve/NXToUFVUVEiS/vrXv2rp0qU6dOiQevTooTlz5mjYsGH%2B2//mN7/RmjVrdOzYMaWmpmrx4sVKSEiwa3xb/fWvf1VBQYEqKioUFham%2B%2B%2B/X3l5eQoPD2cfDSKwAAAADOM1WAAAAIYRWAAAAIYRWAAAAIYRWAAAAIYRWAAAAIYRWAAAAIYRWAAAAIYRWAAAAIYRWAAAAIYRWAAAAIYRWAAAAIYRWAAAAIYRWAAAAIb9f34bVnjlYahAAAAAAElFTkSuQmCC\"/>\n",
       "        </div>\n",
       "        <div role=\"tabpanel\" class=\"tab-pane col-md-12\" id=\"common3896701726749123897\">\n",
       "            \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">534</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">270</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">168</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">169</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">170</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">171</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">172</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">173</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">174</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">175</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"other\">\n",
       "        <td class=\"fillremaining\">Other values (471)</td>\n",
       "        <td class=\"number\">471</td>\n",
       "        <td class=\"number\">97.9%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "        </div>\n",
       "        <div role=\"tabpanel\" class=\"tab-pane col-md-12\"  id=\"extreme3896701726749123897\">\n",
       "            <p class=\"h4\">Minimum 5 values</p>\n",
       "            \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">0</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">1</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">2</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">3</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">5</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "            <p class=\"h4\">Maximum 5 values</p>\n",
       "            \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">530</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">531</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">532</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">533</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">534</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "        </div>\n",
       "    </div>\n",
       "</div>\n",
       "</div><div class=\"row variablerow\">\n",
       "    <div class=\"col-md-3 namecol\">\n",
       "        <p class=\"h4 pp-anchor\" id=\"pp_var_length\">length<br/>\n",
       "            <small>Numeric</small>\n",
       "        </p>\n",
       "    </div><div class=\"col-md-6\">\n",
       "    <div class=\"row\">\n",
       "        <div class=\"col-sm-6\">\n",
       "            <table class=\"stats \">\n",
       "                <tr>\n",
       "                    <th>Distinct count</th>\n",
       "                    <td>477</td>\n",
       "                </tr>\n",
       "                <tr>\n",
       "                    <th>Unique (%)</th>\n",
       "                    <td>99.2%</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Missing (%)</th>\n",
       "                    <td>0.0%</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Missing (n)</th>\n",
       "                    <td>0</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Infinite (%)</th>\n",
       "                    <td>0.0%</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Infinite (n)</th>\n",
       "                    <td>0</td>\n",
       "                </tr>\n",
       "            </table>\n",
       "\n",
       "        </div>\n",
       "        <div class=\"col-sm-6\">\n",
       "            <table class=\"stats \">\n",
       "\n",
       "                <tr>\n",
       "                    <th>Mean</th>\n",
       "                    <td>42969</td>\n",
       "                </tr>\n",
       "                <tr>\n",
       "                    <th>Minimum</th>\n",
       "                    <td>24976</td>\n",
       "                </tr>\n",
       "                <tr>\n",
       "                    <th>Maximum</th>\n",
       "                    <td>74605</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Zeros (%)</th>\n",
       "                    <td>0.0%</td>\n",
       "                </tr>\n",
       "            </table>\n",
       "        </div>\n",
       "    </div>\n",
       "</div>\n",
       "<div class=\"col-md-3 collapse in\" id=\"minihistogram-4137530268564546891\">\n",
       "    <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAMgAAABLCAYAAAA1fMjoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD%2BnaQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAATRJREFUeJzt3MFtwkAARcEQURJF0BPn9EQR6WnTQPRkUDZezMwdaS%2BPLxvj0xhjfAC/%2Btz7ALCy894H%2BAuX2/3hz3x/XSechKOxIBAOsSDPsDpsYUEgLLcgz3yzwywWBIJAIAgEwnLXICt79PrIXa/XZ0EgCASCQCAIBIJAIAgEgtu8E/3XYzNuJ89jQSAIBIJAIAgEgkAgCASCQCAIBIIfCt%2BQN7psZ0EgWJAD8CaYeSwIBIFAEAgEgUAQCASBQBAIBIFAEAgEgUAQCASBQBAIBIFA8Lg7m7zrn6wsCAQLwjRHWB0LAkEgEAQC4TTGGHsfAlZlQSAIBIJAIAgEgkAgCASCQCAIBIJAIAgEgkAgCASCQCAIBIJAIAgEgkAgCASCQCAIBIJAIAgEgkAgCASCQCD8AKZWIG0u3/reAAAAAElFTkSuQmCC\">\n",
       "\n",
       "</div>\n",
       "<div class=\"col-md-12 text-right\">\n",
       "    <a role=\"button\" data-toggle=\"collapse\" data-target=\"#descriptives-4137530268564546891,#minihistogram-4137530268564546891\"\n",
       "       aria-expanded=\"false\" aria-controls=\"collapseExample\">\n",
       "        Toggle details\n",
       "    </a>\n",
       "</div>\n",
       "<div class=\"row collapse col-md-12\" id=\"descriptives-4137530268564546891\">\n",
       "    <ul class=\"nav nav-tabs\" role=\"tablist\">\n",
       "        <li role=\"presentation\" class=\"active\"><a href=\"#quantiles-4137530268564546891\"\n",
       "                                                  aria-controls=\"quantiles-4137530268564546891\" role=\"tab\"\n",
       "                                                  data-toggle=\"tab\">Statistics</a></li>\n",
       "        <li role=\"presentation\"><a href=\"#histogram-4137530268564546891\" aria-controls=\"histogram-4137530268564546891\"\n",
       "                                   role=\"tab\" data-toggle=\"tab\">Histogram</a></li>\n",
       "        <li role=\"presentation\"><a href=\"#common-4137530268564546891\" aria-controls=\"common-4137530268564546891\"\n",
       "                                   role=\"tab\" data-toggle=\"tab\">Common Values</a></li>\n",
       "        <li role=\"presentation\"><a href=\"#extreme-4137530268564546891\" aria-controls=\"extreme-4137530268564546891\"\n",
       "                                   role=\"tab\" data-toggle=\"tab\">Extreme Values</a></li>\n",
       "\n",
       "    </ul>\n",
       "\n",
       "    <div class=\"tab-content\">\n",
       "        <div role=\"tabpanel\" class=\"tab-pane active row\" id=\"quantiles-4137530268564546891\">\n",
       "            <div class=\"col-md-4 col-md-offset-1\">\n",
       "                <p class=\"h4\">Quantile statistics</p>\n",
       "                <table class=\"stats indent\">\n",
       "                    <tr>\n",
       "                        <th>Minimum</th>\n",
       "                        <td>24976</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>5-th percentile</th>\n",
       "                        <td>26872</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Q1</th>\n",
       "                        <td>33032</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Median</th>\n",
       "                        <td>41443</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Q3</th>\n",
       "                        <td>51416</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>95-th percentile</th>\n",
       "                        <td>63154</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Maximum</th>\n",
       "                        <td>74605</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Range</th>\n",
       "                        <td>49629</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Interquartile range</th>\n",
       "                        <td>18384</td>\n",
       "                    </tr>\n",
       "                </table>\n",
       "            </div>\n",
       "            <div class=\"col-md-4 col-md-offset-2\">\n",
       "                <p class=\"h4\">Descriptive statistics</p>\n",
       "                <table class=\"stats indent\">\n",
       "                    <tr>\n",
       "                        <th>Standard deviation</th>\n",
       "                        <td>11695</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Coef of variation</th>\n",
       "                        <td>0.27218</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Kurtosis</th>\n",
       "                        <td>-0.72893</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Mean</th>\n",
       "                        <td>42969</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>MAD</th>\n",
       "                        <td>9828.9</td>\n",
       "                    </tr>\n",
       "                    <tr class=\"\">\n",
       "                        <th>Skewness</th>\n",
       "                        <td>0.43769</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Sum</th>\n",
       "                        <td>20668228</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Variance</th>\n",
       "                        <td>136780000</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Memory size</th>\n",
       "                        <td>3.8 KiB</td>\n",
       "                    </tr>\n",
       "                </table>\n",
       "            </div>\n",
       "        </div>\n",
       "        <div role=\"tabpanel\" class=\"tab-pane col-md-8 col-md-offset-2\" id=\"histogram-4137530268564546891\">\n",
       "            <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAlgAAAGQCAYAAAByNR6YAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD%2BnaQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3X90VPWd//FXkoEIgZAESFDEDZUFDAm/IRrQaFR%2BFlQIBFyPpSoGiEQoKCigcmiBnpBVS1gMbaEqbo1hbfmh/Pgqa6putq1IaxLAKj%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%2BvVKSkr6wY/pdp80%2BAycKTQ0RDExETpxokr19UH3sgs4%2BhtY9Dew6G9gBXN/O3Zsa8vjBuURLLfbLbfbrTvuuEMtW7ZUdHS0br/9du3du1ebN29WfHy80tPTFR4erpSUFKWlpamwsNDush0vNDREISEhCg0NsbuUZon%2BBhb9DSz6G1j0t%2Bm57C7ADnFxcbruuutUUFCghx9%2BWKdPn9aOHTt08803q6ysTAkJCX77JyQkaOvWrY2ev6KiQm6322/M5Wqt2NhYI/U7VVhYqN9PmEV/A4v%2BBhb9DSz62/SCMmCFhoZq5cqVmjJlip5//nlJ0uDBgzVnzhzNmDFDcXFxfvtHRUXJ4/E0ev6CggLl5eX5jWVlZSk7O/vSi28GIiNb2V1Cs0Z/A4v%2BBhb9DSz623SCMmCdOXNG06ZN04gRIzRt2jR9%2BeWXWrx4sebOnWtk/oyMDKWlpfmNuVyt5fFUGZnfqcLCQhUZ2UqVldWqq6u3u5xmh/4GFv0NLPobWMHc3%2BjoCFseNygDVnFxsQ4fPqyf/exnCgsLU9u2bZWdna077rhDN954o7xer9/%2BHo9HMTExjZ4/Nja2wXKg231StbXB9aI%2Bn7q6enoRQPQ3sOhvYNHfwKK/TScoF2Pr6upUX1%2Bvb3%2BA8syZM5KklJQUlZaW%2Bu1fWlqqPn36NGmNAADAuYIyYPXr10%2BtW7fWypUrVV1dLY/Ho9WrV2vQoEG64447VF5ersLCQtXU1KioqEhFRUWaOHGi3WUDAACHCMqAFR0drd/%2B9rd6//33ddNNN%2BnHP/6xrrjiCuXm5qp9%2B/bKz8/X%2BvXrNWDAAC1dulQ5OTnq2bOn3WUDAACHCMoLjdqBC41KLleooqMj5PFUcQ5AANDfwKK/gUV/AyuY%2B8uFRgEAAJoJAhYAAIBhBCwAAADDCFgAAACGEbAAAAAMC8orucMet6942%2B4SLsrWWUPsLgEA4FAcwQIAADCMgAUAAGAYAQsAAMAwAhYAAIBhBCwAAADDCFgAAACGEbAAAAAM4zpYwHmMfOZdu0u4KP9v7o12lwAA%2BD8cwQIAADCMgAUAAGAYAQsAAMAwAhYAAIBhBCwAAADDCFgAAACGEbAAAAAMI2ABAAAYRsACAAAwjIAFAABgGAELAADAMAIWAACAYQQsAAAAwwhYAAAAhrnsLsAOf/3rX3Xffff5jVmWpbNnz%2BrDDz9UcXGxcnNzdeDAAV155ZXKzMzU2LFjbaoWAAA4TVAGrEGDBqmkpMRv7LnnntO%2BfftUUVGhGTNmaMGCBRozZox27dql6dOnq2vXrkpKSrKp4vMb%2Bcy7dpcAAADOwRKhpM8%2B%2B0zr1q3To48%2Bqs2bNys%2BPl7p6ekKDw9XSkqK0tLSVFhYaHeZAADAIQhYkp599lmNHz9eV111lcrKypSQkOC3PSEhQaWlpTZVBwAAnCYolwi/7fDhw9qxY4d27NghSfJ6vYqLi/PbJyoqSh6Pp9FzVlRUyO12%2B425XK0VGxt76QUD5xEWFur3E2bR38Civ4FFf5te0Aesl156ScOGDVPHjh2NzVlQUKC8vDy/saysLGVnZxt7DOBckZGt/H4iMOhvYNHfwKK/TSfoA9b27ds1b9483%2B3o6Gh5vV6/fTwej2JiYho9Z0ZGhtLS0vzGXK7W8niqLq1Y4HtUVlYrMrKVKiurVVdXb3c5zU5YWCj9DSD6G1jB3N/o6AhbHjeoA9bevXtVXl6uIUOG%2BMaSkpL0X//1X377lZaWqk%2BfPo2eNzY2tsFyoNt9UrW1wfWiRtP6%2Bk2zrq6e11oA0d/Aor%2BBRX%2BbTlAvxu7Zs0dRUVFq06aNb2zMmDEqLy9XYWGhampqVFRUpKKiIk2cONHGSgEAgJMEdcA6duxYg3Ov2rdvr/z8fK1fv14DBgzQ0qVLlZOTo549e9pUJQAAcJqgXiLMzMxUZmZmg/FBgwZp48aNNlQEAACag6A%2BggUAABAIBCwAAADDCFgAAACGEbAAAAAMI2ABAAAYRsACAAAwjIAFAABgGAELAADAMAIWAACAYQQsAAAAwwhYAAAAhhGwAAAADCNgAQAAGEbAAgAAMIyABQAAYBgBCwAAwDACFgAAgGEELAAAAMNcdhcAwIzbV7xtdwmNtnXWELtLAICA4ggWAACAYQQsAAAAwwhYAAAAhhGwAAAADCNgAQAAGEbAAgAAMIyABQAAYBgBCwAAwDACFgAAgGEELAAAAMOCOmCtXr1aQ4cOVd%2B%2BfTVlyhQdPnxYklRcXKz09HT1799fo0eP1qZNm2yuFAAAOEnQBqyXXnpJmzZt0gsvvKB33nlH3bp10%2B9%2B9ztVVFRoxowZmjRpkoqLi7VgwQItWrRIJSUldpcMAAAcImi/7Hnt2rWaN2%2BefvSjH0mSFi5cKEn67W9/q/j4eKWnp0uSUlJSlJaWpsLCQiUlJdlWLwAAcI6gPIL1%2Beef6/Dhw/riiy80atQoJScnKzs7WydOnFBZWZkSEhL89k9ISFBpaalN1QIAAKcJyiNYR48elSRt27ZN69atk2VZys7O1sKFC3X69GnFxcX57R8VFSWPx9Po%2BSsqKuR2u/3GXK7Wio2NvfTigWbA5XLe73ZhYaF%2BP2EW/Q0s%2Btv0gjJgWZYlSXrggQd8YWrmzJmaOnWqUlJSLnn%2BgoIC5eXl%2BY1lZWUpOzv7kucGmoPo6Ai7S/jBIiNb2V1Cs0Z/A4v%2BNp2gDFgdOnSQJEVGRvrGOnfuLMuydPbsWXm9Xr/9PR6PYmJiGj1/RkaG0tLS/MZcrtbyeKouoWqg%2BXDiv4WwsFBFRrZSZWW16urq7S6n2aG/gRXM/bXrF7qgDFidOnVSmzZttHfvXvXq1UuSVF5erhYtWig1NVUbN27027%2B0tFR9%2BvRp9PyxsbENlgPd7pOqrQ2uFzVwPk7%2Bt1BXV%2B/o%2Bi939Dew6G/TCcrFWJfLpfT0dD333HP65JNPdPz4ca1atUpjxozRXXfdpfLychUWFqqmpkZFRUUqKirSxIkT7S4bAAA4RFAewZKkOXPm6MyZM5owYYLOnj2r4cOHa%2BHChYqIiFB%2Bfr5%2B/vOfa/HixercubNycnLUs2dPu0sGAAAOEWJ9fcY3AsrtPhmQeUc%2B825A5gUCaeusIXaXcNFcrlBFR0fI46liiSUA6G9gBXN/O3Zsa8vjBuUSIQAAQCARsAAAAAwjYAEAABhGwAIAADCMgAUAAGAYAQsAAMAwAhYAAIBhBCwAAADDCFgAAACGEbAAAAAMI2ABAAAYRsACAAAwjIAFAABgGAELAADAMAIWAACAYQQsAAAAwwhYAAAAhhGwAAAADCNgAQAAGEbAAgAAMIyABQAAYJjL7gIABJ%2BRz7xrdwkXZeusIXaXAMBhOIIFAABgGAELAADAMAIWAACAYZyDBQAXwDljAC4WR7AAAAAMI2ABAAAYRsACAAAwLCjPwerRo4datGihkJAQ39jEiRO1aNEiFRcXKzc3VwcOHNCVV16pzMxMjR071sZqAQCA0wRlwJKkbdu26eqrr/Ybq6io0IwZM7RgwQKNGTNGu3bt0vTp09W1a1clJSXZVCkAAHAalgi/ZfPmzYqPj1d6errCw8OVkpKitLQ0FRYW2l0aAABwkKA9gpWbm6vdu3fr1KlTGjlypObPn6%2BysjIlJCT47ZeQkKCtW7de1NwVFRVyu91%2BYy5Xa8XGxl5y3QBwIS6X8353DgsL9fsJs%2Bhv0wvKgNW3b1%2BlpKTol7/8pT799FPNmjVLixcvltfrVVxcnN%2B%2BUVFR8ng8FzV/QUGB8vLy/MaysrKUnZ19ybUDwIVER0fYXcIPFhnZyu4SmjX623SCMmAVFBT4/n7ttddq7ty5mj59ugYMGGBk/oyMDKWlpfmNuVyt5fFUGZkfAL6PE99rwsJCFRnZSpWV1aqrq7e7nGYnmPtr1y8cQRmwznX11Verrq5OoaGh8nq9fts8Ho9iYmIuar7Y2NgGy4Fu90nV1gbXixqAPZz8XlNXV%2B/o%2Bi939LfpBN1i7J49e7R8%2BXK/sf3796tly5ZKTU1VaWmp37bS0lL16dOnKUsEAAAOF3QBq3379iooKNCaNWt05swZHTx4UM8%2B%2B6wyMjJ0xx13qLy8XIWFhaqpqVFRUZGKioo0ceJEu8sGAAAO4qiAlZaWpry8PB05cuQHzxEXF6c1a9Zo586dSk5O1qRJk3TjjTfqkUceUfv27ZWfn6/169drwIABWrp0qXJyctSzZ0%2BDzwIAADR3IZZlWXYX0VirVq3Sa6%2B9pk8%2B%2BUQ33HCDJk6cqLS0NLlcl/%2BpZG73yYDMO/KZdwMyLwDn2jpriN0lXDSXK1TR0RHyeKo4RygAgrm/HTu2teVxHXUEKysrS6%2B//rpeeeUV/eu//quWLl2q1NRU5eTk6ODBg3aXBwAAIMlhAetrvXr10rx58/Tf//3fevzxx/XKK69o1KhRuv/%2B%2B/XBBx/YXR4AAAhyjgxYZ8%2Be1euvv66pU6dq3rx5iouL02OPPabrrrtOU6ZM0ebNm%2B0uEQAABLHL/%2BSlb9m/f782bNigP/7xj6qqqtLw4cP1/PPP%2B10gdNCgQXrqqac0ZswYGysFAADBzFEBa/To0eratasyMzN15513KioqqsE%2BqampOnHihA3VAQAAfMVRAeuFF17Q4MGDL7jf3//%2B9yaoBgAA4Ls56hysHj16aNq0aXrjjTd8Y7/73e80derUBl9xAwAAYBdHBaxly5bp5MmT6tatm2/s5ptvVn19fYOvvwEAALCLo5YI33nnHW3evFnR0dG%2Bsfj4eK1YsUI//vGPbawMAADgG446gnX69GmFh4c3GA8NDVV1dbUNFQEAADTkqIA1aNAgLV%2B%2BXF988YVv7PPPP9fixYv9LtUAAABgJ0ctET7%2B%2BOO67777dMMNN6hNmzaqr69XVVWVunTpohdffNHu8gAAACQ5LGB16dJFr732mv70pz/pn//8p0JDQ9W1a1cNHTpUYWFhdpcHAAAgyWEBS5Jatmyp2267ze4yAAAAzstRAevTTz9Vbm6uPvroI50%2BfbrB9jfffNOGqgAAAPw5KmA9/vjjqqio0NChQ9W6dWu7ywEAAPhOjgpYpaWlevPNNxUTE2N3KQAAAOflqMs0tG/fniNXAADgsueogJWZmam8vDxZlmV3KQAAAOflqCXCP/3pT3r//ff16quv6uqrr1ZoqH8%2BfPnll22qDAAA4BuOClht2rTRTTfdZHcZAAAA38tRAWvZsmV2lwAAAHBBjjoHS5IOHDiglStX6rHHHvON7d6928aKAAAA/DkqYBUXF2vs2LHasWOHtmzZIumri4/ee%2B%2B9XGQUAABcNhwVsJ5%2B%2Bmk98sgj2rx5s0JCQiR99f2Ey5cv16pVq2yuDgAA4CuOClj/%2BMc/NHnyZEnyBSxJGjFihPbv329XWQAAAH4cFbDatm37nd9BWFFRoZYtW9pQEQAAQEOOClj9%2B/fX0qVLderUKd/YwYMHNW/ePN1www02VgYAAPANRwWsxx57TLt371ZycrJqamrUv39/jRo1Sl6vV/Pnz/9Bcy5dulQ9evTw3S4uLlZ6err69%2B%2Bv0aNHa9OmTabKBwAAQcJR18Hq1KmTtmzZoqKiIh08eFBXXHGFunbtqiFDhvidk9VYe/fu1caNG323KyoqNGPGDC1YsEBjxozRrl27NH36dHXt2lVJSUkmnwoAAGjGHBWwJKlFixa67bbbLnme%2Bvp6Pfnkk5oyZYqeeeYZSdLmzZsVHx%2Bv9PR0SVJKSorS0tJUWFhIwAIAAI3mqICVlpb2vUeqLuZaWC%2B//LLCw8M1ZswYX8AqKytTQkKC334JCQnaunXrDysYAAAEJUcFrFGjRvkFrLq6Oh08eFAlJSX6yU9%2B0uh5jh07ppUrV%2BrFF1/0G/d6vYqLi/Mbi4qKksfjuag6Kyoq5Ha7/cZcrtaKjY29qHkA4IdwuRx1eq0kKSws1O8nzKK/Tc9RAWvu3LnfOb59%2B3b9%2Bc9/bvQ8y5Yt07hx49StWzcdPnzYVHk%2BBQUFysvL8xvLyspSdna28ccCgHNFR0fYXcIPFhnZyu4SmjX623QcFbDO57bbbtMTTzyhJ5544oL7FhcXa/fu3b6v2vm26Ohoeb1evzGPx6OYmJiLqicjI0NpaWl%2BYy5Xa3k8VRc1DwD8EE58rwkLC1VkZCtVVlarrq7e7nKanWDur12/cDSLgLVnzx5ZltWofTdt2qTjx4/rlltukSTf/ZKTk3Xfffc1CF6lpaXq06fPRdUTGxvbYDnQ7T6p2trgelEDsIeT32vq6uodXf/ljv42HUcFrEmTJjUYq66u1v79%2BzVs2LBGzTF//nw9/PDDvttHjx5VRkaGNm7cqPr6euXn56uwsFBjx47V//7v/6qoqEgFBQXGngMAAGj%2BHBWw4uPjG3yKMDw8XOnp6ZowYUKj5mjXrp3atWvnu11bWyvpq2tsSVJ%2Bfr5%2B/vOfa/HixercubNycnLUs2dPQ88AAAAEgxCrsWtruCRu98mAzDvymXcDMi8A59o6a4jdJVw0lytU0dER8niqWMIKgGDub8eObW15XEcdwfrjH//Y6H3vvPPOAFYCAABwfo4KWAsWLFB9fX2DE9pDQkL8xkJCQghYAADANo4KWL/5zW%2B0du1aTZs2TT169JBlWfrwww/161//Wvfcc4%2BSk5PtLhEAAMBZAWv58uVas2aN39XWBw4cqC5duuj%2B%2B%2B//zmtbAQAANDVHXTP/0KFDfp8A/FpkZKTKy8ttqAgAAKAhRwWszp07a/ny5X7fDVhZWanc3Fxdc801NlYGAADwDUctET7%2B%2BOOaM2eOCgoKFBERodDQUJ06dUpXXHGFVq1aZXd5AICL5LRLzTjxEhiwh6MC1tChQ/XWW2%2BpqKhIR48elWVZiouL04033qi2be25zgUAAMC5HBWwJKlVq1a69dZbdfToUXXp0sXucgAAABpw1DlYp0%2Bf1rx589SvXz%2BNHDlS0lfnYD3wwAOqrKy0uToAAICvOCpg5eTkaO/evVqxYoVCQ78pva6uTitWrLCxMgAAgG84KmBt375dv/rVrzRixAjflz5HRkZq2bJl2rFjh83VAQAAfMVRAauqqkrx8fENxmNiYvTll182fUEAAADfwVEnuV9zzTX685//rOTkZL/vHty2bZuuuuoqGysDgMuH0y59ADRHjgpYd999t2bOnKnx48ervr5e69atU2lpqbZv364FCxbYXR4AAIAkhwWsjIwMuVwurV%2B/XmFhYXruuefUtWtXrVixQiNGjLC7PAAAAEkOC1gnTpzQ%2BPHjNX78eLtLAQAAOC9HneR%2B6623%2Bp17BQAAcDlyVMBKTk7W1q1b7S4DAADgezlqifDKK6/UL37xC61Zs0bXXHONWrRo4bc9NzfXpsoAAAC%2B4aiA9fHHH%2BtHP/qRJMnj8dhcDQAAwHdzRMCaPXu2nn76ab344ou%2BsVWrVikrK8vGqgAAAL6bI87B2rlzZ4OxNWvW2FAJAADAhTkiYH3XJwf5NCEAALhcOSJgff3FzhcaAwAAuBw4ImABAAA4CQELAADAMEd8ivDs2bOaM2fOBce4DhYAALgcOCJgDRgwQBUVFRccAwAAuBw4ImB9%2B/pXpuzbt0/Lli1TaWmpwsPDNXjwYC1YsEAdO3ZUcXGxcnNzdeDAAV155ZXKzMzU2LFjjdcAAACap6A8B%2BvMmTO67777NHjwYBUXF2vLli06fvy4nnrqKVVUVGjGjBmaNGmSiouLtWDBAi1atEglJSV2lw0AABwiKANWdXW1Zs%2BerczMTLVs2VIxMTG6/fbb9dFHH2nz5s2Kj49Xenq6wsPDlZKSorS0NBUWFtpdNgAAcAhHLBGa1q5dO02YMMF3%2B8CBA/rDH/6gkSNHqqysTAkJCX77JyQkaOvWrY2ev6KiQm6322/M5Wqt2NjYSyscAGArl8uZxyXCwkL9fiLwgjJgfa28vFzDhw9XbW2tJk6cqOzsbE2dOlVxcXF%2B%2B0VFRV3Ul0sXFBQoLy/PbywrK0vZ2dlG6gYA2CM6OsLuEi5JZGQru0sIGkEdsDp37qySkhJ98skneuKJJ/Too48amTcjI0NpaWl%2BYy5Xa3k8VUbmBwDYw6nv42FhoYqMbKXKymrV1dXbXU6TsisUB3XAkr76yp34%2BHjNnj1bkyZNUmpqqrxer98%2BHo9HMTExjZ4zNja2wXKg231StbXB9aIGgObG6e/jdXX1jn8OThGUi7HFxcUaPny46uu/eZGFhn7Vit69e6u0tNRv/9LSUvXp06dJawQAAM4VlAErMTFRp06dUk5Ojqqrq3XixAmtXLlSAwcO1OTJk1VeXq7CwkLV1NSoqKhIRUVFmjhxot1lAwAAhwjKgNW2bVutXbtWpaWluv766zV69Gi1bdtW//7v/6727dsrPz9f69ev14ABA7R06VLl5OSoZ8%2BedpcNAAAcIsSyLMvuIoKB230yIPOOfObdgMwLAGho66whdpfwg7hcoYqOjpDHUxV052B17NjWlscNyiNYAAAAgUTAAgAAMIyABQAAYBgBCwAAwDACFgAAgGEELAAAAMMIWAAAAIYRsAAAAAwjYAEAABhGwAIAADCMgAUAAGAYAQsAAMAwAhYAAIBhBCwAAADDCFgAAACGEbAAAAAMI2ABAAAYRsACAAAwjIAFAABgGAELAADAMAIWAACAYQQsAAAAwwhYAAAAhhGwAAAADCNgAQAAGEbAAgAAMIyABQAAYBgBCwAAwLCgDVjl5eXKyspScnKyUlJSNH/%2BfFVWVkqS9u7dq3vuuUcDBgzQsGHDtHbtWpurBQAAThK0AWvatGmKjIzUzp079eqrr%2Bqjjz7SL3/5S50%2BfVqZmZm6/vrr9fbbb%2Bvpp59Wfn6%2BduzYYXfJAADAIYIyYFVWVioxMVFz5sxRRESEOnXqpLvuukvvvfee3nrrLZ09e1bTp09X69at1atXL02YMEEFBQV2lw0AABwiKANWZGSkli1bpg4dOvjGjhw5otjYWJWVlalHjx4KCwvzbUtISFBpaakdpQIAAAdy2V3A5aCkpETr16/X6tWrtXXrVkVGRvptj4qKktfrVX19vUJDL5xJKyoq5Ha7/cZcrtaKjY01WjcAoGm5XM48LhEWFur3E4EX9AFr165dmj59uubMmaOUlBRt3br1O/cLCQlp9JwFBQXKy8vzG8vKylJ2dvYl1QoAsFd0dITdJVySyMhWdpcQNII6YO3cuVOPPPKIFi1apDvvvFOSFBMTo0OHDvnt5/V6FRUV1aijV5KUkZGhtLQ0vzGXq7U8niojdQMA7OHU9/GwsFBFRrZSZWW16urq7S6nSdkVioM2YL3//vuaN2%2Benn32WQ0dOtQ3npiYqN///veqra2Vy/VVe0pKStSnT59Gzx0bG9tgOdDtPqna2uB6UQNAc%2BP09/G6unrHPwenCMrF2NraWi1cuFBz5871C1eSlJqaqjZt2mj16tWqrq7W3//%2Bd23YsEGTJ0%2B2qVoAAOA0IZZlWXYX0dTee%2B89/du//ZtatmzZYNu2bdtUVVWlJ598UqWlperQoYOmTp2qu%2B%2B%2B%2B5Ie0%2B0%2BeUn3P5%2BRz7wbkHkBAA1tnTXE7hJ%2BEJcrVNHREfJ4qoLuCFbHjm1tedygXCIcOHCgPvzww%2B/d5/e//30TVQMAAJqboFwiBAAACCQCFgAAgGEELAAAAMMIWAAAAIYRsAAAAAwjYAEAABhGwAIAADCMgAUAAGAYAQsAAMAwAhYAAIBhBCwAAADDCFgAAACGEbAAAAAMI2ABAAAYRsACAAAwjIAFAABgGAELAADAMAIWAACAYQQsAAAAwwhYAAAAhhGwAAAADCNgAQAAGEbAAgAAMIyABQAAYBgBCwAAwDACFgAAgGEELAAAAMMIWAAAAIYRsAAAAAwL2oD19ttvKyUlRbNnz26w7fXXX9eYMWPUr18/jRs3Tu%2B8844NFQIAAKdy2V2AHX79619rw4YN%2Bpd/%2BZcG2/bu3at58%2BYpLy9P119/vbZv366HHnpI27ZtU6dOnWyoFgAAOE1QHsEKDw8/b8AqLCxUamqqUlNTFR4errFjx6p79%2B7atGmTDZUCAAAnCsojWPfee%2B95t5WVlSk1NdVvLCEhQSUlJY2ev6KiQm6322/M5Wqt2NjYiysUAHBZcbmceVwiLCzU7ycCLygD1vfxer1q166d31i7du308ccfN3qOgoIC5eXl%2BY1lZWUpOzvbSI0AAHvcvuJtu0u4KO/9YoTf7cjIVjZVEnwIWN/BsqxLun9GRobS0tL8xlyu1vJ4qi5pXgAALsbX/%2B%2BEhYUqMrKVKiurVVdXb3NVTSs6OsKWxyVgnSNkbNi/AAALDklEQVQ6Olper9dvzOv1KiYmptFzxMbGNlgOdLtPqrY2uF7UAAB7nfv/Tl1dPf8XNREWY8%2BRmJio0tJSv7GSkhL16dPHpooAAIDTELDOMXHiRP3P//yP3nrrLdXU1GjDhg06dOiQxo4da3dpAADAIYJyiTApKUmSVFtbK0l64403JH11pKp79%2B5asWKFli1bpvLycnXr1k35%2Bfnq2LGjbfUCAABnCcqAdaFLLgwbNkzDhg1romoAAEBzwxIhAACAYQQsAAAAwwhYAAAAhhGwAAAADCNgAQAAGBaUnyIEACAYjHzmXbtLaLSts4bYXYJRHMECAAAwjIAFAABgGAELAADAMAIWAACAYQQsAAAAwwhYAAAAhhGwAAAADCNgAQAAGEbAAgAAMIyABQAAYBgBCwAAwDACFgAAgGEELAAAAMMIWAAAAIYRsAAAAAwjYAEAABhGwAIAADCMgAUAAGAYAQsAAMAwAhYAAIBhBCwAAADDCFgAAACGEbAAAAAMI2B9h/Lycj344INKTk7WLbfcopycHNXX19tdFgAAcAiX3QVcjmbOnKlevXrpjTfe0PHjx5WZmakOHTropz/9qd2lAQAAB%2BAI1jlKSkq0b98%2BzZ07V23btlV8fLymTJmigoICu0sDAAAOwRGsc5SVlalz585q166db6xXr146ePCgTp06pTZt2lxwjoqKCrndbr8xl6u1YmNjjdcLAEBz4HI1r2M%2BBKxzeL1eRUZG%2Bo19HbY8Hk%2BjAlZBQYHy8vL8xh566CHNnDnTXKH/571fjDA%2BZ6BUVFSooKBAGRkZhM0AoL%2BBRX8Di/4GFv1tes0rLhpiWdYl3T8jI0Ovvvqq35%2BMjAxD1TmX2%2B1WXl5eg6N7MIP%2BBhb9DSz6G1j0t%2BlxBOscMTEx8nq9fmNer1chISGKiYlp1ByxsbH8hgAAQBDjCNY5EhMTdeTIEZ04ccI3VlJSom7duikiIsLGygAAgFMQsM6RkJCgpKQk5ebm6tSpU9q/f7/WrVunyZMn210aAABwiLCnnnrqKbuLuNzceOON2rJli5YsWaLXXntN6enpuv/%2B%2BxUSEmJ3aY4XERGhwYMHczQwQOhvYNHfwKK/gUV/m1aIdalndAMAAMAPS4QAAACGEbAAAAAMI2ABAAAYRsACAAAwjIAFAABgGAELAADAMAIWAACAYQQsAAAAwwhYAAAAhhGw0Cj79u3TT37yEw0YMEApKSmaNWuW3G63JKm4uFjp6enq37%2B/Ro8erU2bNvnd94UXXtDw4cPVv39/TZ48WaWlpb5tNTU1euKJJ3TTTTcpOTlZ2dnZ8ng8vu3l5eV68MEHlZycrFtuuUU5OTmqr69vmidtk6VLl6pHjx6%2B2/T30vXo0UOJiYlKSkry/VmyZIkk%2BmvK6tWrNXToUPXt21dTpkzR4cOHJdHfS/XXv/7V73WblJSkxMRE33sE/b2MWcAF1NTUWDfccIOVl5dn1dTUWMePH7fuuecea8aMGdbnn39u9e3b1yosLLROnz5tvfvuu1bv3r2tDz74wLIsy3rzzTetgQMHWn/729%2Bs6upqKz8/3xoyZIhVVVVlWZZlLVu2zBo3bpz12WefWR6Px3rooYeszMxM32Pfdddd1sKFC63Kykrr4MGD1rBhw6y1a9fa0oemsGfPHmvw4MFW9%2B7dLcuy6K8h3bt3tz799NMG4/TXjPXr11sjRoyw9u/fb508edJasmSJtWTJEvobIKtXr7Yefvhh%2BnuZI2Dhgrxer/XKK69YZ8%2Be9Y09//zz1u2332795je/se68806//WfNmmUtWrTIsizLevDBB62lS5f6ttXV1VlDhgyxtmzZYp09e9YaMGCA9cYbb/i2f/zxx1aPHj2so0ePWh988IF13XXXWV6v17f9P//zP63hw4cH6qnaqq6uzpowYYL1H//xH76ARX/NOF/Aor9mpKWlWdu3b28wTn/NKy8vtwYPHmyVl5fT38scS4S4oHbt2mnChAlyuVySpAMHDugPf/iDRo4cqbKyMiUkJPjtn5CQ4DsMfe720NBQXXfddSopKdE///lPnTx5Ur169fJtv/baa3XFFVeorKxMZWVl6ty5s9q1a%2Bfb3qtXLx08eFCnTp0K5FO2xcsvv6zw8HCNGTPGN0Z/zcnNzdXNN9%2BsgQMHatGiRaqqqqK/Bnz%2B%2Bec6fPiwvvjiC40aNcq31HTixAn6GwDPPvusxo8fr6uuuor%2BXuYIWGi08vJyJSYmatSoUUpKSlJ2dra8Xq8iIyP99ouKivKt43u9Xr9/oNJXgc3j8cjr9UpSg/tHRkb6tp%2B77eu5vn2eQHNw7NgxrVy5Uk8%2B%2BaTfOP01o2/fvkpJSdGOHTtUUFCgv/3tb1q8eDH9NeDo0aOSpG3btmndunXauHGjjh49qoULF9Jfww4fPqwdO3bopz/9qSTeHy53BCw0WufOnVVSUqJt27bp0KFDevTRRxt1P8uyfvD2C923uVi2bJnGjRunbt26XfR96e%2BFFRQUaMKECWrZsqWuvfZazZ07V1u2bNHZs2cveF/6%2B/2%2Bfo4PPPCA4uLi1KlTJ82cOVM7d%2B68qPv/kO3B0N9ve%2BmllzRs2DB17Nix0fehv/YhYOGihISEKD4%2BXrNnz9aWLVvkcrl8vwl9zePxKCYmRpIUHR3dYLvX61VMTIxvn3O3f/HFF2rfvr1iYmK%2B874hISG%2B%2BzYHxcXF2r17t7Kyshps%2B67%2B0d9Ld/XVV6uurk6hoaH09xJ16NBBkv%2BRkM6dO8uyLJ09e5b%2BGrR9%2B3alpaX5bvP%2BcHkjYOGCiouLNXz4cL%2BP54aGfvXS6d27t9/HfiWptLRUffr0kSQlJiaqrKzMt62urk579uxRnz591KVLF7Vr185v%2Bz/%2B8Q%2BdOXNGiYmJSkxM1JEjR3TixAnf9pKSEnXr1k0REREBea522LRpk44fP65bbrlFycnJGjdunCQpOTlZ3bt3p7%2BXaM%2BePVq%2BfLnf2P79%2B9WyZUulpqbS30vUqVMntWnTRnv37vWNlZeXq0WLFvTXoL1796q8vFxDhgzxjSUlJdHfy1mTn1YPx6msrLRSUlKs5cuXW19%2B%2BaV1/Phx6/7777fuvvtu69ixY1a/fv2sV155xTp9%2BrT11ltvWb1797b27t1rWZZlFRUVWQMGDLB2795tffnll9bKlSut1NRUq7q62rIsy8rJybHuuusu67PPPrNOnDhhZWZmWjNnzvQ99oQJE6zHH3/cOnnypPXxxx9baWlp1vr1623pQ6B4vV7ryJEjvj%2B7d%2B%2B2unfvbh05csQqLy%2Bnv5fo6NGjVt%2B%2Bfa38/HyrpqbGOnDggDVq1ChryZIlvH4NWbp0qXXrrbdahw4dso4dO2ZlZGRY8%2BfPp78GbdiwwRo8eLDfGP29vBGw0Cj79u2z7rnnHqt3797W9ddfb82aNcs6evSoZVmW9Ze//MUaO3as1atXL2vYsGENPq790ksvWampqVZiYqI1efJk68MPP/Rtq6mpsZ566ilr0KBBVr9%2B/ayf/exnVmVlpW/7kSNHrAceeMDq3bu3lZKSYv3qV7%2By6uvrm%2BZJ2%2BTTTz/1XabBsuivCX/5y1%2BsjIwMq2/fvtbgwYOtZcuWWadPn/Zto7%2BX5tt96Nu3rzVv3jzr1KlTlmXRX1Oee%2B45a/To0Q3G6e/lK8SyOIsNAADAJM7BAgAAMIyABQAAYBgBCwAAwDACFgAAgGEELAAAAMMIWAAAAIYRsAAAAAwjYAEAABhGwAIAADCMgAUAAGAYAQsAAMAwAhYAAIBhBCwAAADD/j8HPCNy04LA5wAAAABJRU5ErkJggg%3D%3D\"/>\n",
       "        </div>\n",
       "        <div role=\"tabpanel\" class=\"tab-pane col-md-12\" id=\"common-4137530268564546891\">\n",
       "            \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">33401</td>\n",
       "        <td class=\"number\">2</td>\n",
       "        <td class=\"number\">0.4%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">28915</td>\n",
       "        <td class=\"number\">2</td>\n",
       "        <td class=\"number\">0.4%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">47396</td>\n",
       "        <td class=\"number\">2</td>\n",
       "        <td class=\"number\">0.4%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">33195</td>\n",
       "        <td class=\"number\">2</td>\n",
       "        <td class=\"number\">0.4%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">42339</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">37226</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">62823</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">26981</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">32100</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">67017</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:1%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"other\">\n",
       "        <td class=\"fillremaining\">Other values (467)</td>\n",
       "        <td class=\"number\">467</td>\n",
       "        <td class=\"number\">97.1%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "        </div>\n",
       "        <div role=\"tabpanel\" class=\"tab-pane col-md-12\"  id=\"extreme-4137530268564546891\">\n",
       "            <p class=\"h4\">Minimum 5 values</p>\n",
       "            \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">24976</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">24981</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">25002</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">25242</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">25344</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "            <p class=\"h4\">Maximum 5 values</p>\n",
       "            \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">71246</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">71768</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">72290</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">72413</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">74605</td>\n",
       "        <td class=\"number\">1</td>\n",
       "        <td class=\"number\">0.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "        </div>\n",
       "    </div>\n",
       "</div>\n",
       "</div><div class=\"row variablerow\">\n",
       "    <div class=\"col-md-3 namecol\">\n",
       "        <p class=\"h4 pp-anchor\" id=\"pp_var_text\">text<br/>\n",
       "            <small>Categorical</small>\n",
       "        </p>\n",
       "    </div><div class=\"col-md-3\">\n",
       "    <table class=\"stats \">\n",
       "        <tr class=\"\">\n",
       "            <th>Distinct count</th>\n",
       "            <td>10</td>\n",
       "        </tr>\n",
       "        <tr>\n",
       "            <th>Unique (%)</th>\n",
       "            <td>2.1%</td>\n",
       "        </tr>\n",
       "        <tr class=\"ignore\">\n",
       "            <th>Missing (%)</th>\n",
       "            <td>0.0%</td>\n",
       "        </tr>\n",
       "        <tr class=\"ignore\">\n",
       "            <th>Missing (n)</th>\n",
       "            <td>0</td>\n",
       "        </tr>\n",
       "    </table>\n",
       "</div>\n",
       "<div class=\"col-md-6 collapse in\" id=\"minifreqtable-3460658111431752074\">\n",
       "    <table class=\"mini freq\">\n",
       "        <tr class=\"\">\n",
       "    <th>a05</th>\n",
       "    <td>\n",
       "        <div class=\"bar\" style=\"width:18%\" data-toggle=\"tooltip\" data-placement=\"right\" data-html=\"true\"\n",
       "             data-delay=500 title=\"Percentage: 11.4%\">\n",
       "            &nbsp;\n",
       "        </div>\n",
       "        55\n",
       "    </td>\n",
       "</tr><tr class=\"\">\n",
       "    <th>a07</th>\n",
       "    <td>\n",
       "        <div class=\"bar\" style=\"width:17%\" data-toggle=\"tooltip\" data-placement=\"right\" data-html=\"true\"\n",
       "             data-delay=500 title=\"Percentage: 11.2%\">\n",
       "            &nbsp;\n",
       "        </div>\n",
       "        54\n",
       "    </td>\n",
       "</tr><tr class=\"\">\n",
       "    <th>b10</th>\n",
       "    <td>\n",
       "        <div class=\"bar\" style=\"width:17%\" data-toggle=\"tooltip\" data-placement=\"right\" data-html=\"true\"\n",
       "             data-delay=500 title=\"Percentage: 10.8%\">\n",
       "            &nbsp;\n",
       "        </div>\n",
       "        52\n",
       "    </td>\n",
       "</tr><tr class=\"other\">\n",
       "    <th>Other values (7)</th>\n",
       "    <td>\n",
       "        <div class=\"bar\" style=\"width:100%\" data-toggle=\"tooltip\" data-placement=\"right\" data-html=\"true\"\n",
       "             data-delay=500 title=\"Percentage: 66.5%\">\n",
       "            320\n",
       "        </div>\n",
       "        \n",
       "    </td>\n",
       "</tr>\n",
       "    </table>\n",
       "</div>\n",
       "<div class=\"col-md-12 text-right\">\n",
       "    <a role=\"button\" data-toggle=\"collapse\" data-target=\"#freqtable-3460658111431752074, #minifreqtable-3460658111431752074\"\n",
       "       aria-expanded=\"true\" aria-controls=\"collapseExample\">\n",
       "        Toggle details\n",
       "    </a>\n",
       "</div>\n",
       "<div class=\"col-md-12 extrapadding collapse\" id=\"freqtable-3460658111431752074\">\n",
       "    \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">a05</td>\n",
       "        <td class=\"number\">55</td>\n",
       "        <td class=\"number\">11.4%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">a07</td>\n",
       "        <td class=\"number\">54</td>\n",
       "        <td class=\"number\">11.2%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:98%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">b10</td>\n",
       "        <td class=\"number\">52</td>\n",
       "        <td class=\"number\">10.8%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:94%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">b01</td>\n",
       "        <td class=\"number\">51</td>\n",
       "        <td class=\"number\">10.6%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:92%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">b09</td>\n",
       "        <td class=\"number\">48</td>\n",
       "        <td class=\"number\">10.0%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:87%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">a02</td>\n",
       "        <td class=\"number\">47</td>\n",
       "        <td class=\"number\">9.8%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:85%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">b02</td>\n",
       "        <td class=\"number\">47</td>\n",
       "        <td class=\"number\">9.8%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:85%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">a01</td>\n",
       "        <td class=\"number\">44</td>\n",
       "        <td class=\"number\">9.1%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:80%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">a04</td>\n",
       "        <td class=\"number\">43</td>\n",
       "        <td class=\"number\">8.9%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:78%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">b03</td>\n",
       "        <td class=\"number\">40</td>\n",
       "        <td class=\"number\">8.3%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:73%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "</div>\n",
       "</div><div class=\"row variablerow ignore\">\n",
       "    <div class=\"col-md-3 namecol\">\n",
       "        <p class=\"h4 pp-anchor\" id=\"pp_var_x\"><s>x</s><br/>\n",
       "            <small>Unsupported</small>\n",
       "        </p>\n",
       "    </div><div class=\"col-md-3\">\n",
       "    <p><em>This variable is an unsupported type, check if it needs cleaning or further analysis</em></p>\n",
       "</div>\n",
       "<div class=\"col-md-6\">\n",
       "    <table class=\"stats \">\n",
       "        <tr>\n",
       "            <th>Unsupported value</th>\n",
       "        </tr>\n",
       "    </table>\n",
       "</div>\n",
       "</div><div class=\"row variablerow\">\n",
       "    <div class=\"col-md-3 namecol\">\n",
       "        <p class=\"h4 pp-anchor\" id=\"pp_var_y\">y<br/>\n",
       "            <small>Numeric</small>\n",
       "        </p>\n",
       "    </div><div class=\"col-md-6\">\n",
       "    <div class=\"row\">\n",
       "        <div class=\"col-sm-6\">\n",
       "            <table class=\"stats \">\n",
       "                <tr>\n",
       "                    <th>Distinct count</th>\n",
       "                    <td>7</td>\n",
       "                </tr>\n",
       "                <tr>\n",
       "                    <th>Unique (%)</th>\n",
       "                    <td>1.5%</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Missing (%)</th>\n",
       "                    <td>0.0%</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Missing (n)</th>\n",
       "                    <td>0</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Infinite (%)</th>\n",
       "                    <td>0.0%</td>\n",
       "                </tr>\n",
       "                <tr class=\"ignore\">\n",
       "                    <th>Infinite (n)</th>\n",
       "                    <td>0</td>\n",
       "                </tr>\n",
       "            </table>\n",
       "\n",
       "        </div>\n",
       "        <div class=\"col-sm-6\">\n",
       "            <table class=\"stats \">\n",
       "\n",
       "                <tr>\n",
       "                    <th>Mean</th>\n",
       "                    <td>2.6071</td>\n",
       "                </tr>\n",
       "                <tr>\n",
       "                    <th>Minimum</th>\n",
       "                    <td>0</td>\n",
       "                </tr>\n",
       "                <tr>\n",
       "                    <th>Maximum</th>\n",
       "                    <td>6</td>\n",
       "                </tr>\n",
       "                <tr class=\"alert\">\n",
       "                    <th>Zeros (%)</th>\n",
       "                    <td>25.6%</td>\n",
       "                </tr>\n",
       "            </table>\n",
       "        </div>\n",
       "    </div>\n",
       "</div>\n",
       "<div class=\"col-md-3 collapse in\" id=\"minihistogram-3389567407320575927\">\n",
       "    <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAMgAAABLCAYAAAA1fMjoAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD%2BnaQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAASNJREFUeJzt28EJwkAARUEVS7IIe/JsTxZhT%2Btd5GGEkCXO3IW9vHwTkuMYYxyAj05bHwBmdt76AO8ut8fi3zzv1xVOAhYEkkAgCASCQCAIBIJAIAgEgkAgCASCQCAIBIJAIAgEgkAgCASCQCBM98HUL3xkxVosCIRdLAhz2sOyWxAIFmQHll6pZ7tKz8yCQBAIBIFAEAgEN%2BmT%2BeXRKOuxIBAEAkEgEAQCwU06X/nXhwcCWcArHf/HXywIAoEgEAgCgSAQCAKBIBAIAoEgEAgCgSAQCAKBIBAIAoFwHGOMrQ8Bs7IgEAQCQSAQBAJBIBAEAkEgEAQCQSAQBAJBIBAEAkEgEAQCQSAQBAJBIBAEAkEgEAQCQSAQBAJBIBAEAkEgEF7CZxppvAPH/gAAAABJRU5ErkJggg%3D%3D\">\n",
       "\n",
       "</div>\n",
       "<div class=\"col-md-12 text-right\">\n",
       "    <a role=\"button\" data-toggle=\"collapse\" data-target=\"#descriptives-3389567407320575927,#minihistogram-3389567407320575927\"\n",
       "       aria-expanded=\"false\" aria-controls=\"collapseExample\">\n",
       "        Toggle details\n",
       "    </a>\n",
       "</div>\n",
       "<div class=\"row collapse col-md-12\" id=\"descriptives-3389567407320575927\">\n",
       "    <ul class=\"nav nav-tabs\" role=\"tablist\">\n",
       "        <li role=\"presentation\" class=\"active\"><a href=\"#quantiles-3389567407320575927\"\n",
       "                                                  aria-controls=\"quantiles-3389567407320575927\" role=\"tab\"\n",
       "                                                  data-toggle=\"tab\">Statistics</a></li>\n",
       "        <li role=\"presentation\"><a href=\"#histogram-3389567407320575927\" aria-controls=\"histogram-3389567407320575927\"\n",
       "                                   role=\"tab\" data-toggle=\"tab\">Histogram</a></li>\n",
       "        <li role=\"presentation\"><a href=\"#common-3389567407320575927\" aria-controls=\"common-3389567407320575927\"\n",
       "                                   role=\"tab\" data-toggle=\"tab\">Common Values</a></li>\n",
       "        <li role=\"presentation\"><a href=\"#extreme-3389567407320575927\" aria-controls=\"extreme-3389567407320575927\"\n",
       "                                   role=\"tab\" data-toggle=\"tab\">Extreme Values</a></li>\n",
       "\n",
       "    </ul>\n",
       "\n",
       "    <div class=\"tab-content\">\n",
       "        <div role=\"tabpanel\" class=\"tab-pane active row\" id=\"quantiles-3389567407320575927\">\n",
       "            <div class=\"col-md-4 col-md-offset-1\">\n",
       "                <p class=\"h4\">Quantile statistics</p>\n",
       "                <table class=\"stats indent\">\n",
       "                    <tr>\n",
       "                        <th>Minimum</th>\n",
       "                        <td>0</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>5-th percentile</th>\n",
       "                        <td>0</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Q1</th>\n",
       "                        <td>0</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Median</th>\n",
       "                        <td>3</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Q3</th>\n",
       "                        <td>4</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>95-th percentile</th>\n",
       "                        <td>6</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Maximum</th>\n",
       "                        <td>6</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Range</th>\n",
       "                        <td>6</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Interquartile range</th>\n",
       "                        <td>4</td>\n",
       "                    </tr>\n",
       "                </table>\n",
       "            </div>\n",
       "            <div class=\"col-md-4 col-md-offset-2\">\n",
       "                <p class=\"h4\">Descriptive statistics</p>\n",
       "                <table class=\"stats indent\">\n",
       "                    <tr>\n",
       "                        <th>Standard deviation</th>\n",
       "                        <td>2.1855</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Coef of variation</th>\n",
       "                        <td>0.83831</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Kurtosis</th>\n",
       "                        <td>-1.3833</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Mean</th>\n",
       "                        <td>2.6071</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>MAD</th>\n",
       "                        <td>1.9572</td>\n",
       "                    </tr>\n",
       "                    <tr class=\"\">\n",
       "                        <th>Skewness</th>\n",
       "                        <td>0.23063</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Sum</th>\n",
       "                        <td>1254</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Variance</th>\n",
       "                        <td>4.7765</td>\n",
       "                    </tr>\n",
       "                    <tr>\n",
       "                        <th>Memory size</th>\n",
       "                        <td>3.8 KiB</td>\n",
       "                    </tr>\n",
       "                </table>\n",
       "            </div>\n",
       "        </div>\n",
       "        <div role=\"tabpanel\" class=\"tab-pane col-md-8 col-md-offset-2\" id=\"histogram-3389567407320575927\">\n",
       "            <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAlgAAAGQCAYAAAByNR6YAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD%2BnaQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzt3X90VPWd//FXkiGBJCQkakIX0WBRNCQLAjFtRIGAQrGgYNjIbhdZUX4YSYmkC/LjAKeU0ANULaEItmBRj46w1kAQ5CBKscbugr8mEVmMsJUcMNOQMYSGQn58//BL6hjMBPJJ7p2Z5%2BMc/sjnjve%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%2BXH/605%2BUk5OjG264QUOGDNG%2Bffu0du1a/fa3v1W/fv20ZcsWzZw5U3v27FFkZKQFzwYAAPibgDyDFRER8Z0Fy%2BPxaMaMGRo1apQcDoeGDRumm266SQcPHpQkOZ1OTZw4UQMGDFDXrl318MMPS5LeeuutTn0OAADAfwVkwZoyZYq6d%2B9%2ByW133nmncnJymr%2Bur6%2BX2%2B1WYmKiJKmsrEzJycnN20NDQ3XLLbfI5XJ1bGgAABAwAvIS4eVYvXq1IiMjNXbsWElfn%2BGKjY31ekxsbKyqq6vbvM/Kykq53W6vNYcjUgkJCe0P/A1hYf7Vjx2Ozs97cUb%2BNqvOwnx8Y0a%2BMaPWMR/fAnFGQVuwmpqatHr1ahUXF2vLli2KiIjw2tYeTqdThYWFXms5OTnKzc1t1379XVxclGXHjonpZtmx/QHz8Y0Z%2BcaMWsd8fAukGQVlwWpsbNQTTzyhjz/%2BWC%2B99JJ69%2B7dvC0uLk4ej8fr8R6PRzfeeGOb95%2Bdna3MzEyvNYcjUtXVZ9sX/Fv8rembfv5tERYWqpiYbqqpqVNDQ2OnH9/umI9vzMg3ZtQ65uNbR87Iqn/cB2XBWrFihY4ePaqXXnpJPXr08NqWkpKisrIyTZgwQZLU0NCgTz75RFlZWW3ef0JCQovLgW73GdXXB/cby8rn39DQGPTzbw3z8Y0Z%2BcaMWsd8fAukGfnXKRADDh06pO3bt2vjxo0typUkTZ48Wa%2B99po%2B/PBD1dXVaf369QoPD9fw4cM7PywAAPBLAXkGKzU1VdLXnxCUpL1790qSXC6X/uu//ktnzpzRiBEjvP6btLQ0bdq0SXfeeacef/xxzZkzR1VVVUpNTdXGjRvVtWvXzn0SAADAb4U0tfeObrSJ233G%2BD4djlDdtfqA8f12FCt%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%2B/HjddNNN2r59uyTJ6XRq6tSp%2Bv73v6/o6Gjl5eWpvLxcH330UWc/DQAA4KccVgfoCFOmTPnObWVlZRo2bJjXWnJyslwul86dO6fPPvtMycnJzduio6N1/fXXy%2BVyaeDAgW06fmVlpdxut9eawxGphISEy3gWvoWF%2BVc/djg6P%2B/FGfnbrDoL8/GNGfnGjFrHfHwLxBkFZMFqjcfjUWxsrNdabGysPvvsM3311Vdqamq65Pbq6uo2H8PpdKqwsNBrLScnR7m5uVcePADExUVZduyYmG6WHdsfMB/fmJFvzKh1zMe3QJpR0BUsST7vp2rv/VbZ2dnKzMz0WnM4IlVdfbZd%2B/02f2v6pp9/W4SFhSompptqaurU0NDY6ce3O%2BbjGzPyjRm1jvn41pEzsuof90FXsOLi4uTxeLzWPB6P4uPj1aNHD4WGhl5y%2B1VXXdXmYyQkJLS4HOh2n1F9fXC/sax8/g0NjUE//9YwH9%2BYkW/MqHXMx7dAmpF/nQIxICUlRaWlpV5rLpdLAwYMUEREhG688UaVlZU1b6upqdFf/vIX/fM//3NnRwUAAH4q6ArWv/zLv%2Bjdd9/V22%2B/rb///e/atm2bjh8/rvHjx0uSJk%2BerC1btqi8vFy1tbVavXq1brnlFqWmplqcHAAA%2BIuAvER4sQzV19dLkvbu3Svp6zNVN910k1avXq2CggJVVFSob9%2B%2B2rBhg6655hpJ0gMPPCC3261///d/19mzZ5Went7ihnUAAIDWBGTBcrlcrW6/%2B%2B67dffdd19yW0hIiHJzc4P%2BE38AAODKBd0lQgAAgI5GwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYUFbsD755BNNmTJFQ4YM0e233678/HydPn1aklRSUqKsrCwNGjRI99xzj7Zv325xWgAA4E%2BCsmDV19dr%2BvTpGjhwoN59910VFxfr9OnTWrp0qSorK/Xoo4/qgQceUElJiRYuXKjFixfL5XJZHRsAAPiJoCxYbrdbbrdb9957r8LDwxUXF6e77rpLhw8f1o4dO5SUlKSsrCxFREQoIyNDmZmZ2rp1q9WxAQCAnwjKgpWYmKhbbrlFTqdTZ8%2BeVVVVlfbs2aPhw4errKxMycnJXo9PTk5WaWmpRWkBAIC/CcqCFRoaqrVr1%2BrNN9/UoEGDlJGRofr6es2dO1cej0cxMTFej%2B/Ro4eqq6stSgsAAPyNw%2BoAVjh//rxmzpypMWPGaObMmfrb3/6mZcuWKT8/38j%2BKysr5Xa7vdYcjkglJCQY2f9FYWH%2B1Y8djs7Pe3FG/jarzsJ8fGNGvjGj1jEf3wJxRkFZsEpKSnTixAk9/vjjCgsLU/fu3ZWbm6t7771Xd9xxhzwej9fjq6urFR8f3%2Bb9O51OFRYWeq3l5OQoNzfXSH5/FRcXZdmxY2K6WXZsf8B8fGNGvjGj1jEf3wJpRkFZsBoaGtTY2KimpqbmtfPnz0uSMjIy9Ic//MHr8aWlpRowYECb95%2Bdna3MzEyvNYcjUtXVZ9uRuiV/a/qmn39bhIWFKiamm2pq6tTQ0Njpx7c75uMbM/KNGbWO%2BfjWkTOy6h/3QVmwbr31VkVGRmrt2rWaOXOmzp07p/Xr1ystLU333nuvCgsLtXXrVo0fP17vvfee9u/fL6fT2eb9JyQktLgc6HafUX19cL%2Bx7lp9wOoIl2XXnNutjtBpGhoag/716Qsz8o0ZtY75%2BBZIM/KvUyCGxMXF6Xe/%2B53ef/993Xnnnfrxj3%2Bsrl27as2aNbrqqqu0YcMGvfDCCxo8eLBWrFihVatW6eabb7Y6NgAA8BNBeQZLklJSUvT8889fcltaWpqKioo6OREAAAgUtjuDlZmZqcLCQp08edLqKAAAAFfEdgXr/vvv1%2Buvv65Ro0bp4Ycf1p49e1RfX291LAAAgDazXcHKycnR66%2B/rldeeUU33nijVqxYoWHDhmnVqlU6duyY1fEAAAB8sl3Buqh///6aN2%2Be3nrrLS1YsECvvPKKxo4dq2nTpunjjz%2B2Oh4AAMB3sm3BunDhgl5//XU98sgjmjdvnhITE/XEE0/olltu0dSpU7Vjxw6rIwIAAFyS7T5FWF5erm3btum1117T2bNnNXr0aP3%2B97/X4MGDmx%2BTlpampUuXaty4cRYmBQAAuDTbFax77rlHffr00YwZM3TfffepR48eLR4zbNgwnT592oJ0AAAAvtmuYG3ZskW33Xabz8d99NFHnZAGAADg8tnuHqx%2B/fpp5syZ2rt3b/Pac889p0ceeaTFL2EGAACwI9sVrIKCAp05c0Z9%2B/ZtXhs%2BfLgaGxu1cuVKC5MBAAC0je0uEb7zzjvasWOH4uLimteSkpK0evVq/fjHP7YwGQAA/uVHT/3J6ghtdvAXY6yOYJTtzmCdO3dOERERLdZDQ0NVV1dnQSIAAIDLY7uClZaWppUrV%2Bqrr75qXvvyyy%2B1bNkyrx/VAAAAYFe2u0S4YMECPfTQQ/rhD3%2Bo6OhoNTY26uzZs%2Brdu7eef/55q%2BMBAAD4ZLuC1bt3b%2B3cuVN//OMf9Ze//EWhoaHq06ePhg4dqrCwMKvjAQAA%2BGS7giVJ4eHhGjVqlNUxAAAArojtCtYXX3yhNWvW6OjRozp37lyL7W%2B%2B%2BaYFqQAAANrOdgVrwYIFqqys1NChQxUZGWl1HAAAgMtmu4JVWlqqN998U/Hx8VZHAQAAuCK2%2BzENV111FWeuAACAX7NdwZoxY4YKCwvV1NRkdRQAAIArYrtLhH/84x/1/vvv69VXX9W1116r0FDvDvjyyy9blAwAAKBtbFewoqOjdeedd1odAwAA4IrZrmAVFBRYHQEAAKBdbHcPliR9/vnnWrt2rZ544onmtQ8%2B%2BMDCRAAAAG1nuzNYJSUleuSRR9SnTx8dP35cBQUF%2BuKLLzRlyhQ99dRTGjlypNURAcDWfvTUn6yO0Ga75txudQSgQ9juDNaTTz6pn/3sZ9qxY4dCQkIkff37CVeuXKl169ZZnA4AAMA32xWs//3f/9XkyZMlqblgSdKYMWNUXl5uVSwAAIA2s13B6t69%2ByV/B2FlZaXCw8MtSAQAAHB5bFewBg0apBUrVqi2trZ57dixY5o3b55%2B%2BMMfWpgMAACgbWx3k/sTTzyhBx98UOnp6WpoaNCgQYNUV1enG2%2B8UStXrrQ6HgAAgE%2B2K1g9e/ZUcXGx9u/fr2PHjqlr167q06ePbr/9dq97sgAAAOzKdgVLkrp06aJRo0ZZHQMAAOCK2K5gZWZmtnqm6s033%2BzENAAAAJfPdgVr7NixXgWroaFBx44dk8vl0oMPPmhhMgAAgLaxXcHKz8%2B/5Pobb7yhP//5z52cBgAA4PLZ7sc0fJdRo0Zp586dVscAAADwyW8K1ieffKKmpiarYwAAAPhku0uEDzzwQIu1uro6lZeX6%2B6777YgEQAAwOWxXcFKSkpq8SnCiIgIZWVladKkSRalAgAAaDvbFazO/Gnt69ev14svvqja2loNHDhQy5cv17XXXquSkhKtWbNGn3/%2Bub73ve9pxowZGj9%2BfKflAgAA/s12Beu1115r82Pvu%2B%2B%2BKz7Oiy%2B%2BqO3bt2vLli1KSEjQU089peeee07Tp0/Xo48%2BqoULF2rcuHE6dOiQZs2apT59%2Big1NfWKjwcAAIKH7QrWwoUL1djY2OKG9pCQEK%2B1kJCQdhWsTZs2ad68ebrhhhskSYsWLZIk/e53v1NSUpKysrIkSRkZGcrMzNTWrVspWAAAoE1sV7B%2B%2B9vfatOmTZo5c6b69eunpqYmHTlyRM8%2B%2B6x%2B8pOfKD09vd3H%2BPLLL3XixAl99dVXGjt2rKqqqpSenq6lS5eqrKxMycnJXo9PTk7Wrl272rz/yspKud1urzWHI1IJCQntzv5NYWF%2B8yFQv%2BRwBP58L76GeC19N2bUsXif4ZsCaUa2K1grV67Uxo0blZiY2Lw2ZMgQ9e7dW9OmTVNxcXG7j3Hq1ClJ0u7du7V582Y1NTUpNzdXixYt0rlz57yOLUk9evRQdXV1m/fvdDpVWFjotZaTk6Pc3Nx2Z0fniYuLsjpCp4mJ6WZ1BNtjRh2D9xm%2BKZBmZLuCdfz4ccXGxrZYj4mJUUVFhZFjXLzU%2BPDDDzeXqdmzZ%2BuRRx5RRkZGu/efnZ2tzMxMrzWHI1LV1Wfbve9vCqSmb0emv192FBYWqpiYbqqpqVNDQ6PVcWyJGXUs3mf4po6YkVUl3nYFq1evXlq5cqV%2B%2BtOfKi4uTpJUU1OjX//617ruuuuMHOPqq6%2BW9HVp%2B%2BZxm5qadOHCBXk8Hq/HV1dXKz4%2Bvs37T0hIaHE50O0%2Bo/p63lj%2BJJi%2BXw0NjUH1fK8EM%2BoYwTRTXkO%2BBdKMbFewFixYoLlz58rpdCoqKkqhoaGqra1V165dtW7dOiPH6Nmzp6Kjo3X48GH1799fklRRUaEuXbpo2LBhKioq8np8aWmpBgwYYOTYAAAg8NmuYA0dOlRvv/229u/fr1OnTqmpqUmJiYm644471L17dyPHcDgcysrK0jPPPKO0tDRFR0dr3bp1GjdunCZMmKDf/OY32rp1q8aPH6/33ntP%2B/fvl9PpNHJsAAAQ%2BGxXsCSpW7duGjlypE6dOqXevXt3yDHmzp2r8%2BfPa9KkSbpw4YJGjx6tRYsWKSoqShs2bNDy5cu1bNky9erVS6tWrdLNN9/cITkAAEDgsV3BOnfunJYsWaKdO3dK%2BvryXE1NjR5//HH96le/8rpvqj3Cw8O1ZMkSLVmypMW2tLS0FpcJAQAA2sp2H0NbtWqVDh8%2BrNWrVys09B/xGhoatHr1aguTAQAAtI3tCtYbb7yhX//61xozZkzzL32OiYlRQUGB9uzZY3E6AAAA32xXsM6ePaukpKQW6/Hx8frb3/7W%2BYEAAAAuk%2B3uwbruuuv05z//Wenp6V6/e3D37t36p3/6JwuTAQhWQxbutjoCAD9ju4L1r//6r5o9e7buv/9%2BNTY2avPmzSotLdUbb7yhhQsXWh0PAADAJ9sVrOzsbDkcDr3wwgsKCwvTM888oz59%2Bmj16tUaM2aM1fEAAAB8sl3BOn36tO6//37df//9VkcBAAC4Ira7yX3kyJFe914BAAD4G9sVrPT0dO3atcvqGAAAAFfMdpcIv/e97%2BkXv/iFNm7cqOuuu05dunTx2r5mzRqLkgEAALSN7QrWZ599phtuuEGSVF1dbXEaAACAy2ebgpWXl6cnn3xSzz//fPPaunXrlJOTY2EqAACAy2ebe7D27dvXYm3jxo0WJAEAAGgf2xSsS31ykE8TAgAAf2SbgnXxFzv7WgMAALA72xQsAACAQEHBAgAAMMw2nyK8cOGC5s6d63ONn4MFAADszjYFa/DgwaqsrPS5BgAAYHe2KVjf/PlXAAAA/ox7sAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwxxWBwAAwF/86Kk/WR0BfoIzWAAAAIZRsAAAAAwL%2BoK1YsUK9evXr/nrkpISZWVladCgQbrnnnu0fft2C9MBAAB/FNT3YB0%2BfFhFRUXNX1dWVurRRx/VwoULNW7cOB06dEizZs1Snz59lJqaamFSAADgT4L2DFZjY6OWLFmiqVOnNq/t2LFDSUlJysrKUkREhDIyMpSZmamtW7daFxQAAPidoC1YL7/8siIiIjRu3LjmtbKyMiUnJ3s9Ljk5WaWlpZ0dDwAA%2BLGgvET417/%2BVWvXrtXzzz/vte7xeJSYmOi11qNHD1VXV1/W/isrK%2BV2u73WHI5IJSQkXFng7xAWFrT9uFM4HIE/34uvIV5L343ZdKxgeJ%2Bh7QLp/RaUBaugoEATJ05U3759deLECeP7dzqdKiws9FrLyclRbm6u8WOh48TFRVkdodPExHSzOgKCVDC9z%2BBbIP1dFHQFq6SkRB988IGKi4tbbIuLi5PH4/Faq66uVnx8/GUdIzs7W5mZmV5rDkekqqvPXn7gVgRS07cj098vOwoLC1VMTDfV1NSpoaHR6ji2xPusYwXD%2Bwxt1xF/F1lV4oOuYG3fvl1VVVUaMWKEJKmpqUmSlJ6eroceeqhF8SotLdWAAQMu6xgJCQktLge63WdUX8//wPxJMH2/Ghoag%2Br5wj543eGbAunvoqArWPPnz9dPf/rT5q9PnTql7OxsFRUVqbGxURs2bNDWrVs1fvx4vffee9q/f7%2BcTqeFiQEAgL8JuoIVGxur2NjY5q/r6%2BslST179pQkbdiwQcuXL9eyZcvUq1cvrVq1SjfffLMlWQEAgH8KuoL1bddee62OHDnS/HVaWprXDx8F/IU//RLaXXNutzoCAHQo7t4EAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAwL2oJVUVGhnJwcpaenKyMjQ/Pnz1dNTY0k6fDhw/rJT36iwYMH6%2B6779amTZssTgsAAPxJ0BasmTNnKiYmRvv27dOrr76qo0eP6pe//KXOnTunGTNm6Ac/%2BIEOHDigJ598Uhs2bNCePXusjgwAAPxEUBasmpoapaSkaO7cuYqKilLPnj01YcIEHTx4UG%2B//bYuXLigWbNmKTIyUv3799ekSZPkdDqtjg0AAPyEw%2BoAVoiJiVFBQYHX2smTJ5WQkKCysjL169dPYWFhzduSk5O1devWNu%2B/srJSbrfba83hiFRCQkL7gn9LWFhQ9uNO43Aw347iT7Plfdax/Om1gI4XSO%2B3oCxY3%2BZyufTCCy9o/fr12rVrl2JiYry29%2BjRQx6PR42NjQoN9f3NdzqdKiws9FrLyclRbm6u0dzoWHFxUVZHCFjMFhfxWsA3xcR0szqCMUFfsA4dOqRZs2Zp7ty5ysjI0K5duy75uJCQkDbvMzs7W5mZmV5rDkekqqvPtivrtwVS07cj098v/IM/zZb3Wcfyp9cCOl5NTZ0aGhqN7tOqEh/UBWvfvn362c9%2BpsWLF%2Bu%2B%2B%2B6TJMXHx%2Bv48eNej/N4POrRo0ebzl5JUkJCQovLgW73GdXXm33RoGPx/eo4zBYX8VrANzU0NAbMayJo/2n2/vvva968eXr66aeby5UkpaSk6MiRI6qvr29ec7lcGjBggBUxAQCAHwrKglVfX69FixYpPz9fQ4cO9do2bNgwRUdHa/369aqrq9NHH32kbdu2afLkyRalBQAA/iYoC9aHH36o8vJyLV%2B%2BXKmpqV5/3G63nnnmGb377ru67bbbNGfOHOXl5Wn48OFWxwYAAH4iKO/BGjJkiI4cOdLqY1566aVOSgMAAAJNUJ7BAgAA6EgULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGEULAAAAMMoWAAAAIZRsAAAAAyjYAEAABhGwQIAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjIIFAABgGAXrEioqKjR9%2BnSlp6drxIgRWrVqlRobG62OBQAA/ITD6gB2NHv2bPXv31979%2B5VVVWVZsyYoauvvlr/8R//YXU0AADgBziD9S0ul0uffvqp8vPz1b17dyUlJWnq1KlyOp1WRwMAAH6CM1jfUlZWpl69eik2NrZ5rX///jp27Jhqa2sVHR3tcx%2BVlZVyu91eaw5HpBISEoxmDQujH3ckh4P5dhR/mi3vs47lT68FdLxAer9RsL7F4/EoJibGa%2B1i2aqurm5TwXI6nSosLPRae%2ByxxzR79mxzQfV1kXuw51FlZ2cbL2%2BBorKyUk6nMyhmdPAXYy77vwmm%2BVwp3me%2BBdPriPdZx6isrNTatWsDakaBUxUlWGcnAAAHGklEQVQNampqatd/n52drVdffdXrT3Z2tqF0/%2BB2u1VYWNjibBn%2BgRm1jvn4xox8Y0atYz6%2BBeKMOIP1LfHx8fJ4PF5rHo9HISEhio%2BPb9M%2BEhISAqaBAwCAy8cZrG9JSUnRyZMndfr06eY1l8ulvn37KioqysJkAADAX1CwviU5OVmpqalas2aNamtrVV5ers2bN2vy5MlWRwMAAH4ibOnSpUutDmE3d9xxh4qLi/Xzn/9cO3fuVFZWlqZNm6aQkBCro7UQFRWl2267jbNrrWBGrWM%2BvjEj35hR65iPb4E2o5Cm9t7RDQAAAC9cIgQAADCMggUAAGAYBQsAAMAwChYAAIBhFCwAAADDKFgAAACGUbAAAAAMo2ABAAAYRsECAAAwjILlhyoqKjR9%2BnSlp6drxIgRWrVqlRobG62OZTsHDhxQRkaG8vLyrI5iSxUVFcrJyVF6eroyMjI0f/581dTUWB3LVj799FM9%2BOCDGjx4sDIyMjRnzhy53W6rY9nSihUr1K9fP6tj2E6/fv2UkpKi1NTU5j8///nPrY5lO%2BvXr9fQoUM1cOBATZ06VSdOnLA6UrtRsPzQ7NmzlZiYqL1792rz5s3au3evfv/731sdy1aeffZZLV%2B%2BXNdff73VUWxr5syZiomJ0b59%2B/Tqq6/q6NGj%2BuUvf2l1LNs4f/68HnroId12220qKSlRcXGxqqqqxK9vbenw4cMqKiqyOoZt7d69Wy6Xq/nP4sWLrY5kKy%2B%2B%2BKK2b9%2BuLVu26J133lHfvn313HPPWR2r3ShYfsblcunTTz9Vfn6%2BunfvrqSkJE2dOlVOp9PqaLYSERGhbdu2UbC%2BQ01NjVJSUjR37lxFRUWpZ8%2BemjBhgg4ePGh1NNuoq6tTXl6eZsyYofDwcMXHx%2Buuu%2B7S0aNHrY5mK42NjVqyZImmTp1qdRT4qU2bNikvL0833HCDoqOjtWjRIi1atMjqWO1GwfIzZWVl6tWrl2JjY5vX%2Bvfvr2PHjqm2ttbCZPYyZcoUde/e3eoYthUTE6OCggJdffXVzWsnT55UQkKChansJTY2VpMmTZLD4ZAkff755/rDH/6gH/3oRxYns5eXX35ZERERGjdunNVRbGvNmjUaPny4hgwZosWLF%2Bvs2bNWR7KNL7/8UidOnNBXX32lsWPHKj09Xbm5uTp9%2BrTV0dqNguVnPB6PYmJivNYulq3q6morIiEAuFwuvfDCC5o1a5bVUWynoqJCKSkpGjt2rFJTU5Wbm2t1JNv461//qrVr12rJkiVWR7GtgQMHKiMjQ3v27JHT6dSHH36oZcuWWR3LNk6dOiXp68uomzdvVlFRkU6dOsUZLFijqanJ6ggIIIcOHdK0adM0d%2B5cZWRkWB3Hdnr16iWXy6Xdu3fr%2BPHj%2Bs///E%2BrI9lGQUGBJk6cqL59%2B1odxbacTqcmTZqk8PBwff/731d%2Bfr6Ki4t1/vx5q6PZwsX/nz388MNKTExUz549NXv2bO3bt09///vfLU7XPhQsPxMfHy%2BPx%2BO15vF4FBISovj4eItSwV/t27dP06dP14IFCzRlyhSr49hWSEiIkpKSlJeXp%2BLi4oC4fNFeJSUl%2BuCDD5STk2N1FL9y7bXXqqGhQVVVVVZHsYWLtyl888pMr1691NTU5PczomD5mZSUFJ08edLrL3iXy6W%2BffsqKirKwmTwN%2B%2B//77mzZunp59%2BWvfdd5/VcWynpKREo0eP9voRKKGhX/%2BV2aVLF6ti2cb27dtVVVWlESNGKD09XRMnTpQkpaena%2BfOnRans4dPPvlEK1eu9ForLy9XeHg49zv%2Bfz179lR0dLQOHz7cvFZRUaEuXbr4/YwoWH4mOTlZqampWrNmjWpra1VeXq7Nmzdr8uTJVkeDH6mvr9eiRYuUn5%2BvoUOHWh3HllJSUlRbW6tVq1aprq5Op0%2Bf1tq1azVkyBA%2BQCFp/vz5euONN1RUVKSioiJt3LhRklRUVKTMzEyL09nDVVddJafTqY0bN%2Br8%2BfM6duyYnn76aWVnZyssLMzqeLbgcDiUlZWlZ555Rv/3f/%2BnqqoqrVu3TuPGjWv%2BgIm/Cmnihh6/c%2BrUKS1evFj//d//rejoaD3wwAN67LHHFBISYnU020hNTZX0dZGQ1PxGdblclmWyk4MHD%2Brf/u3fFB4e3mLb7t271atXLwtS2c%2BRI0e0fPlyffzxx4qMjNQPfvADzZ8/X4mJiVZHs50TJ05o5MiROnLkiNVRbOV//ud/tGbNGh05ckTh4eGaMGGC8vLyFBERYXU02zh//rwKCgq0c%2BdOXbhwQaNHj9bixYv9/qoMBQsAAMAwLhECAAAYRsECAAAwjIIFAABgGAULAADAMAoWAACAYRQsAAAAwyhYAAAAhlGwAAAADKNgAQAAGEbBAgAAMIyCBQAAYBgFCwAAwDAKFgAAgGH/D2D/BkNBIrI4AAAAAElFTkSuQmCC\"/>\n",
       "        </div>\n",
       "        <div role=\"tabpanel\" class=\"tab-pane col-md-12\" id=\"common-3389567407320575927\">\n",
       "            \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">0</td>\n",
       "        <td class=\"number\">123</td>\n",
       "        <td class=\"number\">25.6%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">1</td>\n",
       "        <td class=\"number\">79</td>\n",
       "        <td class=\"number\">16.4%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:64%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">6</td>\n",
       "        <td class=\"number\">74</td>\n",
       "        <td class=\"number\">15.4%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:60%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">4</td>\n",
       "        <td class=\"number\">68</td>\n",
       "        <td class=\"number\">14.1%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:55%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">3</td>\n",
       "        <td class=\"number\">56</td>\n",
       "        <td class=\"number\">11.6%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:46%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">5</td>\n",
       "        <td class=\"number\">43</td>\n",
       "        <td class=\"number\">8.9%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:35%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">2</td>\n",
       "        <td class=\"number\">38</td>\n",
       "        <td class=\"number\">7.9%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:31%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "        </div>\n",
       "        <div role=\"tabpanel\" class=\"tab-pane col-md-12\"  id=\"extreme-3389567407320575927\">\n",
       "            <p class=\"h4\">Minimum 5 values</p>\n",
       "            \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">0</td>\n",
       "        <td class=\"number\">123</td>\n",
       "        <td class=\"number\">25.6%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">1</td>\n",
       "        <td class=\"number\">79</td>\n",
       "        <td class=\"number\">16.4%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:64%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">2</td>\n",
       "        <td class=\"number\">38</td>\n",
       "        <td class=\"number\">7.9%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:31%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">3</td>\n",
       "        <td class=\"number\">56</td>\n",
       "        <td class=\"number\">11.6%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:46%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">4</td>\n",
       "        <td class=\"number\">68</td>\n",
       "        <td class=\"number\">14.1%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:55%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "            <p class=\"h4\">Maximum 5 values</p>\n",
       "            \n",
       "<table class=\"freq table table-hover\">\n",
       "    <thead>\n",
       "    <tr>\n",
       "        <td class=\"fillremaining\">Value</td>\n",
       "        <td class=\"number\">Count</td>\n",
       "        <td class=\"number\">Frequency (%)</td>\n",
       "        <td style=\"min-width:200px\">&nbsp;</td>\n",
       "    </tr>\n",
       "    </thead>\n",
       "    <tr class=\"\">\n",
       "        <td class=\"fillremaining\">2</td>\n",
       "        <td class=\"number\">38</td>\n",
       "        <td class=\"number\">7.9%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:51%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">3</td>\n",
       "        <td class=\"number\">56</td>\n",
       "        <td class=\"number\">11.6%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:75%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">4</td>\n",
       "        <td class=\"number\">68</td>\n",
       "        <td class=\"number\">14.1%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:91%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">5</td>\n",
       "        <td class=\"number\">43</td>\n",
       "        <td class=\"number\">8.9%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:58%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr><tr class=\"\">\n",
       "        <td class=\"fillremaining\">6</td>\n",
       "        <td class=\"number\">74</td>\n",
       "        <td class=\"number\">15.4%</td>\n",
       "        <td>\n",
       "            <div class=\"bar\" style=\"width:100%\">&nbsp;</div>\n",
       "        </td>\n",
       "</tr>\n",
       "</table>\n",
       "        </div>\n",
       "    </div>\n",
       "</div>\n",
       "</div>\n",
       "    <div class=\"row headerrow highlight\">\n",
       "        <h1>Correlations</h1>\n",
       "    </div>\n",
       "    <div class=\"row variablerow\">\n",
       "    <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAm0AAAIiCAYAAACaBf8HAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD%2BnaQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3XlYlXX%2B//EX54CobIJy0EzTVNx1NM0196KshrLMZTJN21xQSTNbLDXLqTCd0kqb0XSmGposdcj5mdhe9p0cTZGRLLRcUsHkyOLCdn5/MJ48iXLw5vacO56P6zoX8rmX8%2BbjAd687uUEuFwulwAAAODXbL4uAAAAABWjaQMAALAAmjYAAAALoGkDAACwAJo2AAAAC6BpAwAAsACaNgAAAAugaQMAALAAmjYAAAALoGkDAACwgEBfFwDg0mrZsuV5lwUFBalOnTpq166dbrnlFsXFxSkgIOASVgcAOJ8A3nsUqF7ONG0dOnRQdHS0x7LTp09rz549%2BumnnyRJ/fv314svvqgaNWpc8joBAJ5o2oBq5kzTtmTJEg0aNKjcdVJTU/XQQw/pxIkTGjt2rB5%2B%2BOFLWSIAoByc0wbgHIMGDdK0adMkSW%2B99ZZOnz7t44oAAJzTBqBc1113nZ566imdPHlS//3vf9WpUyeP5Zs3b9Ybb7yhb775Rk6nU6GhoWrVqpVuu%2B023XzzzeXu0%2Bl0asWKFfrkk0/0448/qrCwUBEREerQoYPuvvtudevW7ZxtBgwYoIMHD%2Bq1117T6dOntXDhQu3fv1%2Bvv/66rrrqKknS7t27tXz5cm3ZskVHjhxRYGCgHA6Hrr76at15553lnseXl5enlStXatOmTfrxxx9VVFSkevXqqXPnzhozZozat2/vsf6BAwc0cOBASdKOHTuUmZmpV155Rdu2bZPT6ZTD4dDAgQOVmJio2rVrX9ScA8CFkLQBKFedOnXc/87NzfVYtnDhQo0ZM0YbN25UaGiounXrpqioKG3evFnTp0/X1KlTVVJS4rFNdna2hgwZoldffVV79uxRmzZt1L17d9WoUUMfffSRRo8erdWrV5%2B3nj179igxMVFBQUHq3r27atasKUn697//rdtuu03vvfeeTp8%2BrauuukqdOnVSfn6%2B3n77bd1xxx3avHmzx74OHjyoW2%2B9VS%2B99JK%2B//57tWrVSt27d5fL5VJKSoruuOMOvfPOO%2BetZevWrRo5cqTS09PVqlUrNW3aVAcPHtSqVauUkJDg9RwDQKW4AFQrsbGxrtjYWNfGjRsvuF5mZqZ73R07drjHP/jgA1dsbKyra9eurv/7v//z2Obf//63q2fPnq7Y2FjXihUrPJbNmzfPFRsb64qLi3MdPXrUPV5cXOx6%2BumnXbGxsa6rrrrKlZ%2Bf77Fd//79XbGxsa4BAwa4XnnllXPqHDp0qCs2Ntb1xz/%2B0VVSUuKx34ULF7piY2NdN910k8c2I0eOdMXGxrqGDBniOnLkiHu8pKTE9dJLL7liY2Nd7dq1c%2B3Zs8e9bP/%2B/e756Nevn2vZsmWu0tJS9/L33nvPvXzXrl0XmloAuCgkbQDK9cEHH0iSIiIi1Lp1a/f4K6%2B8Ikl6%2BOGHdfXVV3ts07VrV82YMUOStHLlSo9lDodDN954oyZNmqS6deu6x%2B12uxITE2Wz2ZSXl6dvvvmm3HpKS0t13333nTO%2Ba9cuSdKQIUNks/3yI81ut2vy5MmaPHmyRo0apcLCQknS9u3btWXLFknSc889J4fD4d7GZrNp0qRJatu2rQoLC/X3v/%2B93FpatGihe%2B%2B91%2BN2KPHx8YqKinI/BwBUNc5pA3COTz75REuXLpUk3XPPPQoMLPtRceTIEaWnp0uSrr322nK3HTRokAICAvTTTz9p7969atq0qSTp3nvvPe/z1apVS3Xr1lV2drays7PLXad79%2B4eTdkZ4eHhOnr0qNavX68pU6Z4LLPZbJo4caLH2GeffSap7CraZs2anfdrSE9PP%2Bew6hnlnbMXEBCgRo0a6dixY8rJySl3OwAwgqYNqKaWLl2qd99912OssLBQP/zwg/bv3y%2BpLL2655573Mt3797t/vfMmTPPu%2B/AwEAVFRXphx9%2BcDdtZ/b/ySefaOfOncrKylJubq5c/7vrUF5enqSyRK08MTEx5Y6PHj1aCxYs0Msvv6zPPvtMN910k3r06HHemwh///33ksrSsvO58sorJZWdR1eexo0blzseHBwsSSoqKjrvvgHgYtG0AdXUjh07zhmz2%2B2KjIzUwIEDNWzYMPXt29dj%2BfHjx93/3rRpU4XPcaYRk6SMjAxNnDhRBw4cuKh6IyIiyh0/c8h06dKlSktLU1pamiQpOjpa119/vUaPHq1GjRqdU1N4ePh5nyssLExSWfN16tQp90UPZ3CzYQC%2BQNMGVFMXurnu%2BZw5hysoKEhpaWlev8XVqVOnNH78eP30009q0qSJJkyYoB49eigyMlJBQUGSfrm1R0XPXZ777rtPw4cPV2pqqj755BN9%2BeWXys7O1l//%2BlclJyfrueee0w033OD11%2Bk6657j5R2SBQBfoGkD4LUztwEpKipSTk6O%2B8T7inz66af66aefFBAQoGXLlumKK644Z51Tp04Zqi08PFxDhgzRkCFDVFJSoi%2B//FJLlizRtm3b9Nhjj6l79%2B6KjIx0J3a/vo3J2c4sq1mzJqkaAL/Bn5AAvBYbG%2Bv%2B93fffef1dj/88IOksnPBymvY9u3bp59//tlwfWfY7XZdc801WrVqlWJiYlRQUOA%2BHHzmXLazz8/7tTPLLnTeGwBcajRtALwWHR2ttm3bSpLefvvtctfZu3ev4uPj3bcGkX45R%2Bx8b4e1ZMkS979/fVPeC9m%2BfbseffRRLV%2B%2BvNzlNWrUUL169ST9kuSdOU9v9%2B7d5TZupaWl7tudXHPNNV7XAgBmo2kDUCkPPPCAJCklJUWvv/66x/lfP/74oyZPnqyMjAwdOnTIPd6qVStJ0uHDh5WamuoeP3nypObNm6e0tDT322Sd74rN8tjtdq1evVovvviiPv7443OWb9y4Ubt27VJQUJA6d%2B4sSWrTpo169%2B4tSXr00Ud17Ngx9/rFxcV6/vnnlZmZqbCwMA0fPtzrWgDAbJzTBqBSrrvuOt13331atmyZ5s%2Bfr1WrVunKK69UTk6Odu3apZKSErVp00bTp093b9OpUyf17t1bn3/%2BuRISEtSxY0cFBwdr586dCgwM1IoVK/T%2B%2B%2B9r27ZtWrVqlb799luNHz9eXbt2vWAt7dq109ixY7V8%2BXLdf//9atiwoa644grZbDYdOHDAfVj2scceU3R0tHu7Z555RqNHj1ZaWpoGDBig2NhYBQcH67vvvlNOTo5q1qypBQsWnPc2IwDgCzRtACpt2rRp6tmzp/72t7/pm2%2B%2B0ebNmxUcHKx27dpp8ODBGjly5Dkn8C9atEhJSUnatGmTdu7cKYfDobi4OD3wwANq3Lix6tevr//%2B97/aunWrvvvuO6%2BvTH344YfVtWtXrVmzRjt37tS2bdtUXFys6Oho3XTTTbrzzjvPebP7mJgYvfPOO3r99de1ceNGfffddyouLlZMTIyuu%2B46jRs3rtxz7wDAlwJcZx/bAAAAgF/inDYAAAALoGkDAACwAJo2AAAAC6BpAwAA1dZnn32mnj17KjEx8YLrlZaWauHChRo4cKC6du2qcePGaf/%2B/e7lTqdTU6dOVc%2BePdW7d2899thjht/p5ddo2gAAQLX02muvad68eV5dLf7GG2/on//8p5YtW6aPPvpITZo00cSJE933qpw1a5ZOnjyplJQUrV69WpmZmUpKSqrSemnaAABAtRQcHKx33nnHq6YtOTlZY8aMUbNmzRQaGqrExERlZmZq%2B/btOnr0qFJTU5WYmKioqCjFxMRowoQJWr16tYqKiqqsXu7TBgAALCsrK0vZ2dkeY9HR0XI4HBVue9ddd3n1HKdOndL333%2BvNm3auMdCQ0N1xRVXKC0tTXl5ebLb7WrZsqV7edu2bXXixAnt2bPHY9wImjazeHljUJ9q2lT67jupRQtp715fV1Oh407/v6VgQIAUFibl5Un%2BfgfEiNIcX5fgHZtNCg%2BXcnOl0lJfV3NB2cX%2B/yPVZgtQVFSIjh0rUGmpn79IJUVfeZmvS6hYkybSN99Iv/ud9L934fBreXm%2BeV6Tfi8mv/iiFi9e7DE2adIkJSQkVNlzHD9%2BXC6XSxERER7jERERysnJUZ06dRQaGupxU/Az6%2BbkVN3PWv//CQPz1Kkj2e1lH1ElAgJ%2Befh702YZZ08qDLPZAhQQECCbLcASTZslRESU/Sz91S90XBrDhg3TgAEDPMbOftu6qnSh9yO4FO9VQNMGAADMZzPnNHqHw%2BHVoVAj6tSpI5vNJqfT6THudDpVt25dRUVFKT8/XyUlJbLb7e5lklS3bt0qq4MLEQAAgPlsNnMel0BwcLBatGih9PR091hubq727dunDh06qHXr1nK5XMrIyHAvT0tLU3h4uJo2bVplddC0AQAA/MqRI0d0/fXXu%2B/FNmLECK1atUqZmZnKz89XUlKSWrdurfbt2ysqKkpxcXFatGiRjh07psOHD2vJkiW6/fbbFRhYdQc1OTwKAADMd4lSscpo3769JKm4uFiSlJqaKqksJSsqKtLevXtVWFgoSRo%2BfLiys7M1atQoFRQUqFu3bh4XQMydO1dPPvmkBg4cqKCgIN10000V3rC3sgJcl%2BLMuerICidNd%2Bokbd0qde4sbdvm62oqZIWrR222X64e9fMLHa1z9ajd/svVoyUlvq7mgqxw9WhgoE2RkSHKySlQcbGfv0hlkatHO3aUPv9c6t1b2r7d19VUzFdXjwYHm7Pf06fN2a8f8v%2BfMAAAwPr8MGmzGpo2AABgPpo2w5hBAAAACyBpAwAA5iNpM4wZBAAAsACSNgAAYD6SNsNo2gAAgPlo2gxjBgEAACyApA0AAJiPpM0wZhAAAMACSNoAAID5SNoMo2kDAADmo2kzjBkEAACwAJI2AABgPpI2w5hBAAAACyBpAwAA5iNpM4ymDQAAmI%2BmzTBmEAAAwAJI2gAAgPlI2gxjBgEAACyApA0AAJiPpM0wmjYAAGA%2BmjbDmEEAAAALIGkDAADmI2kzjBkEAACwAJI2AABgPpI2w2jaAACA%2BWjaDGMGAQAALICkDQAAmI%2BkzTBmEAAAwAL8umk7ePCg2rdvr71791Z626SkJI0aNcqEqgAAQKXZbOY8qhG/PjzasGFDpaWl%2BboMAABgVDVrsMzADAIAAFiAXzdtBw4cUMuWLZWZmakBAwboH//4h%2B677z516tRJgwYN0ueff%2B5e98MPP1RcXJw6deqkqVOn6tSpUx77Wr9%2BveLj4/W73/1OAwcOVHJysiTp6NGjuvrqq/Xxxx%2B7133wwQc1fvz4S/I1AgBQLXB41DC/Pjz6a3/5y1/03HPPqVWrVpo9e7aeeeYZrV%2B/Xrm5uUpMTNT06dM1bNgwbd68WdOmTVPr1q0lSWlpaXrsscf00ksvqUePHtq2bZvuvfdetWjRQp07d9ZDDz2k%2BfPnq2fPntqxY4c%2B/fRTvf/%2B%2B17XlZWVpezsbI%2Bx6KZN5ahTp0q//irXqpXnRz9nhe/NMzVaoVYF2H1dgXcsNKmB/v13sCTJbrd5fPR7HTv6uoKKxcZ6fvRn27f7ugIYYKmmrX///urQoYMkKS4uTmvWrFFpaak%2B//xz1a5dW3/4wx9ks9nUt29fdenSRQUFBZKkd999V/369VPv3r0lSV26dNENN9ygtWvXqnPnzrr99tu1bt06LV%2B%2BXBs2bNBDDz2kmJgYr%2BtKTk7W4sWLPcYmTZmihClTqugrN9mbb/q6Aq%2BE%2BbqASggJ8XUF3gj3dQGVExrq6woqFOnrAiohPLyWr0vwzllHVPze8uW%2BrqBiYT78SWqBP7z8naWatssvv9z975o1a6qkpERFRUU6fPiwGjRoINtZL4gmTZooPT1dkrRv3z5t3rxZ7du3dy93uVzuJi4gIEDz5s3TzTffrHbt2umOO%2B6oVF3Dhg3TgAEDPMaib75ZWrmy0l/jJdWqVVnDNnKklJHh62oqlPfJVl%2BXUCGbraxhKyiQSkt9Xc2FhblyfV2Cd2y2soYtP9/vJzWnxP/TS7vdpvDwWsrNPamSEv%2BeT0mKvDnO1yVULDa2rGEbO1bavdvX1fgvmjbDLNW02c7zH15YWKiSkhKPsdKzfrjXrFlTI0aM0KxZs86778OHDyswMFCHDx/WyZMnVbt2ba/rcjgccjgcnoMXcZsSn8nIkLZt83UVFfLz39ceSkstUG9pScXr%2BJPSUqnEv2suLg7wdQleKykpVXGxv79IZa3Debt3W6teWM5vou11OBw6cuSIXC6XeywzM9P978aNG%2Bvbb7/12Obw4cPuRq%2BwsFBPPPGE5syZoxYtWmjRokWXpnAAAKoLLkQw7Dfx1fbs2VP5%2Bfn6%2B9//rsLCQqWmpmr7WX/t3H777dq6datWr16twsJC7dq1S0OHDtWGDRskSUuWLJHD4dDNN9%2Bsxx57TG%2B//bbH9gAAwCCaNsN%2BE19t/fr1tWDBAi1fvlxXX3211q1bp5EjR7qXN2vWTAsWLNCf//xndenSRQkJCRo3bpwGDx6s3bt3a%2BXKlXryyScllaVyY8eO1eOPP66ioiJffUkAAAAeAlxnH1NE1QmwwLktnTpJW7dKnTtb4py2407/f6nabGUXZ%2BXl%2Bf85bRGlOb4uwTt2uxQeLuXm%2Bv05bdnF/n%2BacGCgTZGRIcrJKbDEOW3RV17m6xIq1rFj2VWuvXtb45y2vDzfPG%2BvXubs94svzNmvH/L/nzAAAAAmOXjwoObMmaPt27erdu3aGjx4sKZNm3bOxY9jx47V119/7TFWXFysiRMnatKkSRo1apS2bt3qsV3Tpk21bt26KquVpg0AAJjPT88/S0hIUNu2bZWamqqff/5Z999/v%2BrVq6e7777bY73lv7oPX25urgYPHqxrr73WPfbUU09pyJAhptXqnzMIAAB%2BW/zwQoS0tDRlZGRo%2BvTpCgsLU5MmTTRmzBj3W11eyKJFi3TttdeqZcuWhmqoDJI2AABgWeW%2BlWR09Ln3Ty1Henq6GjZsqIiICPdY27ZttXfvXuXn5yv0PO/E8uOPP2rNmjVKTU31GF%2B/fr3%2B/Oc/69ChQ%2BrYsaPmzp2rxo0bX8RXVT6SNgAAYD6Tkrbk5GQNGTLE4%2BFNUiZJTqdT4eGeb%2Bl3poHLyTn/xVrLli3TbbfdpqioKPdYs2bN1KJFC7355pvatGmToqKidM8996iwsPAiJqt8JG0AAMCyyn0ryehor7ev7E00nE6n1q5dq3/9618e47Nnz/b4fO7cuerWrZv%2B85//qEePHpV6jvOhaQMAAOYz6UKEct9K0ktRUVFyOp0eY06nUwEBAR4p2tk2bdqkpk2bqlGjRhfcd2hoqCIiInTkyJGLqq08HB4FAADm88MLEdq1a6dDhw7p2LFj7rG0tDQ1b95cISEh5W6zadMm9frVPefy8/M1e/Zsjwbt2LFjOnbsWIXNXWXQtAEAgGqpTZs2at%2B%2BvRYsWKD8/HxlZmZqxYoVGjFihCTp%2Buuv15YtWzy22bVrly6//HKPsdDQUG3fvl3z5s2T0%2BnU8ePHNWfOHLVs2VKdOnWqsnpp2gAAgPn8MGmTpBdffFFZWVnq1auX7rrrLt1yyy3ut8Lcu3evTpw44bF%2Bdna26tWrd85%2BlixZIpfLpbi4OPXr109FRUVatmzZOTfpNYJz2gAAQLVVv359vfbaa%2BUu%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%2B/btPR5Hjx6VJJ0%2BfVpPPPGE%2BvTpo27dumny5MnKycmp0lpp2gAAgPlsNnMeBiUkJCgmJkapqalasWKFUlNTtXLlynLXjY%2BPV1pamsejXr16kqSFCxcqPT1dycnJ2rBhg1wulx555BHD9Z2Npg0AAJjPD5u2tLQ0ZWRkaPr06QoLC1OTJk00ZswYJScnV2o/xcXFeueddzRhwgQ1aNBAderU0dSpU/Xxxx/ryJEjhmo8G%2Be0AQAAy8rKylJ2drbHWHR0tBwOR4Xbpqenq2HDhoqIiHCPtW3bVnv37lV%2Bfr5CQ0M91v/22281fPhw7d69Ww0aNNAjjzyi3r17a9%2B%2BfcrLy1Pbtm3d6zZr1kw1a9ZUenq6YmJiDH6VZWjaAACA%2BUy6ECE5OVmLFy/2GJs0aZISEhIq3NbpdCo8PNxj7EwDl5OT49G01a9fX40aNdK0adPkcDiUnJysBx54QOvWrZPT6ZSkc/YVHh5epee10bQBAADLGjZsmAYMGOAxFh0d7fX2LpfLq/WGDh2qoUOHuj8fM2aM3n//fa1bt059%2BvSp1L4uFk0bAAAwn0lJm8Ph8OpQaHmioqLcKdkZTqdTAQEBioqKqnD7hg0bKisry72u0%2BlUSEiIe/nx48dVt27di6qtPFyIAAAAzOeHFyK0a9dOhw4d0rFjx9xjaWlpat68uUfzJUkvv/yyNm/e7DGWmZmpRo0aqVGjRoqIiFB6erp72e7du1VYWKh27doZqvFsNG0AAKBaatOmjdq3b68FCxYoPz9fmZmZWrFihUaMGCFJuv7667VlyxZJZSnanDlztGfPHp0%2BfVrLly/Xvn37dOutt8put%2BuOO%2B7Qq6%2B%2BqkOHDiknJ0cvvPCCrr32WvctQaoCh0cBAID5/PQdEV588UXNmjVLvXr1UmhoqIYPH66RI0dKkvbu3asTJ05IkqZNmyap7Fw2p9Op5s2b6/XXX1f9%2BvUlSZMnT1ZBQYHi4%2BNVXFys/v37a/bs2VVaa4DL7LPmqqnjx31dQcVsNiksTMrLk8q5%2BbPfiagT4OsSKtapk7R1q9S5s7Rtm6%2BruaC8XGt869tsUkiIVFDg/6/TU6fyfF1ChQIDbYqMDFFOToGKi/18QiUFBob5uoQK2e1SeLiUmyuVlPi6mopFRvroiefPN2e/VXwDW39G0gYAAMznp0mbldC0AQAA89G0GcYMAgAAWABJGwAAMB9Jm2HMIAAAgAWQtAEAAPORtBlG0wYAAMxH02YYMwgAAGABJG0AAMB8JG2GMYMAAAAWQNIGAADMR9JmGE0bAAAwH02bYdV%2BBkePHq0//vGPHmNLlizR8OHDfVQRAADAuap90nbLLbfohRde0IwZM2T7318BH3zwge644w6v95GVlaXs7GyPsVq1ohUd7ajSWqvamT96LPPHT6dOvq6gYq1aeX70Y1b5f7fS6zQw0P%2BLtNttHh/9nd3u6woqZqXXaEmJD5/cChPk56p903bddddp7ty5%2Br//%2Bz/16NFD%2B/fvV2Zmpm644Qav95GcnKzFixd7jE2cOEmTJydUdbmmCAnxdQVe2rrV1xV47803fV1Bhazy335GrVq%2BrqBiIZb5ZpLCwy0woRYTGurrCiqWk%2BPrCmBEtW/aQkJCNGjQIK1bt049evTQBx98oF69eikqKsrrfQwbNkwDBgzwGKtVK1p5eVVdbdWy2coatoICqbTU19VULKxvZ1%2BXULFWrcoatpEjpYwMX1dzQQWfWaMJttnKGraTJ/3/dVpYWODrEipkt9sUHl5LubknVVLi5xMqyW73/0bYZitr2PLz/f816lMkbYZV%2B6ZNKjtEOnnyZM2ZM0cbN27UnXfeWantHQ6HHA7PQ6HHj1vnm7e01CK1btvm6wq8l5Hh9/Va4v/8LFZ4nRYX%2B3mBZykpKbVUvVZQWurjw4/%2BjqbNMGZQUo8ePRQSEqJ//OMf%2Bu677zRw4EBflwQAAOCBpE2SzWbTzTffrBdeeEEDBw5ULSucPAMAgJWQtBnGDP7PLbfcovz8fN18882%2BLgUAAOAcJG3/8/PPP6thw4bq1auXr0sBAOC3h6TNMJo2ld1n7ZlnntG4cePc92oDAABViN%2BvhlX7GVy6dKluuOEGde3aVSNGjPB1OQAAAOWq9knb/fffr/vvv9/XZQAA8NtG0mZYtW/aAADAJUDTZhgzCAAAYAEkbQAAwHwkbYYxgwAAABZA0gYAAMxH0mYYTRsAADAfTZthzCAAAIAFkLQBAADzkbQZxgwCAABYAEkbAAAwH0mbYTRtAADAfDRthjGDAAAAFkDSBgAAzEfSZhgzCAAAYAEkbQAAwHwkbYbRtAEAAPPRtBnGDAIAAFgASRsAADAfSZthzCAAAKi2Dh48qPvuu0/dunVT//799fzzz6u0tLTcdd966y3FxcWpU6dOio%2BPV2pqqnvZzJkz1aZNG7Vv39796NKlS5XWStIGAADM56dJW0JCgtq2bavU1FT9/PPPuv/%2B%2B1WvXj3dfffdHutt2LBBCxYs0NKlS9WhQwetWbNGU6dO1b/%2B9S81atRIkjR%2B/HglJCSYVqt/ziAAAPhtsdlMeWRlZSk9Pd3jkZWV5VVJaWlpysjI0PTp0xUWFqYmTZpozJgxSk5OPmfdU6dO6cEHH9RVV12loKAgDR06VCEhIfrmm2%2BqeqZsCiv%2BAAAgAElEQVTOi6QNAABYVnJyshYvXuwxNmnSJK8Sr/T0dDVs2FARERHusbZt22rv3r3Kz89XaGioezw%2BPt5j29zcXBUUFCgmJsY99tVXX2nTpk368ccf1axZM82ePVvt2rW72C/tHDRtAADAfCYdHh02bJgGDBjgMRYdHe3Vtk6nU%2BHh4R5jZxq4nJwcj6btbC6XS48//rg6duyoq6%2B%2BWpLUqFEj2Ww2TZkyRSEhIVq8eLHGjh2rDRs2KDIysrJfVrlo2gAAgGU5HA45HI6L3t7lclVq/aKiIs2cOVPff/%2B9Vq1a5R6fOHGix3oPPfSQUlJSlJqaqqFDh150fWejaQMAAObzwwsRoqKi5HQ6PcacTqcCAgIUFRV1zvqnTp3ShAkTdPLkSb3xxhsXTNDsdrsaNGjg9fl13vC/GQQAAL89Jl2IYES7du106NAhHTt2zD2Wlpam5s2bKyQkxGNdl8ulxMREBQYG6vXXX/do2Fwul%2BbPn6%2BMjAz3WGFhofbt2%2Be%2BsrQq0LQBAIBq6cx91RYsWKD8/HxlZmZqxYoVGjFihCTp%2Buuv15YtWyRJ//znP/X999/rT3/6k4KDgz32ExAQoAMHDmjOnDk6cuSICgoKlJSUpKCgIA0aNKjK6uXwKAAAMJ8fHh6VpBdffFGzZs1Sr169FBoaquHDh2vkyJGSpL179%2BrEiROSpNWrV%2BvgwYPuCw/OiI%2BP17x58/T000/r2Wef1ZAhQ5Sfn68OHTpo5cqVql27dpXVGuCq7Bl48Mrx476uoGI2mxQWJuXlSee5%2BbNfiagT4OsSKtapk7R1q9S5s7Rtm6%2BruaC8XGt869tsUkiIVFDg/6/TU6fyfF1ChQIDbYqMDFFOToGKi/18QiUFBob5uoQK2e1SeLiUmyuVlPi6mopV0YWMlbd5szn77dHDnP36IZI2AABgPj9N2qyEpg0AAJiPps0wmjaTRJTm%2BLqEigXYJYUrzJUrlfp/pm%2BFw3k2mxQiqeCzrX5/KC8s3AKHmyX3IeeQa/z/kPOprFxfl/CbE1n7tK9LqFhAgKQaCq9ZKFnijKPgileBX6JpAwAA5iNpM4wZBAAAsACSNgAAYD6SNsNo2gAAgPlo2gxjBgEAACyApA0AAJiPpM0wZhAAAMACSNoAAID5SNoMo2kDAADmo2kzjBkEAACwAJI2AABgPpI2w2jaAACA%2BWjaDGMGAQAALICkDQAAmI%2BkzTBmEAAAwAJI2gAAgPlI2gyjaQMAAOajaTOMGQQAALAAkjYAAGA%2BkjbDmEEAAAALIGkDAADmI2kzjKYNAACYj6bNMGYQAADAAkjaAACA%2BUjaDGMGAQAALICkDQAAmI%2BkzTCaNgAAYD6aNsOYQQAAAAsgaQMAAOYjaTOMGQQAALAAkjYAAGA%2BkjbDaNoAAID5aNoMYwYBAAAsgKQNAACYj6TNMGYQAADAAkjaAACA%2BUjaDKNpAwAA5qNpM4wZBAAAsAC/adoOHDigli1bKjMz85I%2B76hRo5SUlHRJnxMAgGrHZjPnUY1Ur69W0v79%2B/X//t//83UZAADADxw8eFD33XefunXrpv79%2B%2Bv5559XaWlpueuuWrVKcXFx6ty5s0aMGKGdO3e6l50%2BfVpPPPGE%2BvTpo27dumny5MnKycmp0lqrXdP2wQcfaMOGDb4uAwCA6sVPk7aEhATFxMQoNTVVK1asUGpqqlauXHnOeh9%2B%2BKFeeuklPffcc/ryyy/Vv39/PfDAAzpx4oQkaeHChUpPT1dycrI2bNggl8ulRx55xHB9Z/PLCxGcTqfmzZunr776SgUFBerevbtmz56tmJgYHThwQAMHDtTy5cv1/PPP64cfflCrVq2UlJSkyy%2B/XJL08ssva8WKFQoMDNT999%2BvTz75RJ07d1bt2rXdh0JTU1P1zTffSJJKSkr0xBNPKCUlRcHBwZo1a5YGDx7sdb1ZWVnKzs72GIuuWVOO6OgqmhGTnHmxWyRetkKZlprSTp18XYF3WrXy/OjHAgP9/z/ebrd5fPR7AQG%2BrqBiZ2q0Qq0ul%2B%2Be26QfjOX%2BDo6OlsPhqHDbtLQ0ZWRkaMWKFQoLC1NYWJjGjBmjlStX6u677/ZYNzk5WUOGDFHHjh0lSffcc49WrVqljz76SHFxcXrnnXf07LPPqkGDBpKkqVOn6sYbb9SRI0cUExNTJV%2BrXzZtM2fOVGBgoN5//33Z7XY9%2BeSTeuSRR7R8%2BXL3OqtWrdLSpUsVHBysu%2B66S3/%2B8581e/Zsbdy4Ua%2B%2B%2BqqWL1%2BuNm3aaP78%2BUpPT1fnzp01btw4fffddzp9%2BrQWLlzo3ldKSoqeeeYZPf7441q8eLFmz56t6667ToGB3k1PcnKyFi9e7DE2aeJEJUyeXDUTYrbQUF9X4JUQXxdQCbVq%2BboCL2zd6usKKufNN31dQYUifV1AJYSHW%2BFFajFBQb6uoGKnT/u6gipX7u/gSZOUkJBQ4bbp6elq2LChIiIi3GNt27bV3r17lZ%2Bfr9Czfj%2Bmp6d7BDo2m02tW7dWWlqaWrdurby8PLVt29a9vFmzZqpZs6bS09N/u03bsWPH9NFHH2n9%2BvXuSZw%2Bfbr69evn0UmPGDHCPQm9e/dWWlqaJOmTTz5R79691aVLF0nSjBkztHbt2gs%2BZ%2BfOnXXNNddIkq6//notXbpUx44d86pLl6Rhw4ZpwIABHmPRNWtKublebe8zNltZw5afL53n%2BL0/KbCH%2B7qECtlsZQ3byZP%2BP6Uh13T2dQneadWqrGEbOVLKyPB1NReUs%2BkzX5dQIbvdpvDwWsrNPamSEj9/kUqKDLFAIxQQUNawFRX5Nsnycy6Zk0SW%2BzvYyyNdTqdT4eGev1vO9B45OTkeTZvT6fRo7s6sm5OTI6fTKUnn7Cs8PLxKz2vzu6Yt6H9/qdxyyy0e43a7XYcOHVJUVJQkuQ%2BFSlKtWrV0%2Bn9/PWRnZ6tx48buZWFhYWrSpMkFn/PsfQUHB0uSCgsLva7Z4XCc2%2BDl5EglJV7vw6dKSy1Ra6kFjjycUVrq/02btm3zdQWVk5Hh9zUXF/v7f/ovSkpKrVGvlZogl8ta9f5GlPs7uBJclfg/q2jdyuzrYvhd01azZk1J0qeffqrIyHMPNhw4cECSFHCecwdKS0vPOaxpq%2BA4%2Bvn2BQAAqoZZf8ja7Re/bVRUlDslO8PpdCogIMAdEp0RGRlZ7rotWrRwr%2Bt0OhUS8svJPMePH1fdunUvvsBf8cszUW02m7799lv350VFRTpy5IhX29atW1c//fST%2B/P8/Hzt3bu3ymsEAADeO3MEoqofRrRr106HDh3SsWPH3GNpaWlq3ry5R/N1Zt309HT35yUlJfrvf/%2Brjh07qlGjRoqIiPBYvnv3bhUWFqpdu3bGijyL3zVtQUFBGjx4sJKSknT48GGdOnVKL7zwgsaOHetV7Ni9e3d9%2Bumn2rFjh06dOqXnnnvOnd5JZYc/Dx06pNzcXBUXF5v5pQAAAD/Wpk0btW/fXgsWLFB%2Bfr4yMzO1YsUKjRgxQlLZee5btmyRVHYu/Zo1a/TNN9/o5MmTeuWVV1SjRg3169dPdrtdd9xxh1599VUdOnRIOTk5euGFF3TttdeqXr16VVav3zVtkjRr1ixdccUVuvHGG3XNNdfo%2B%2B%2B/18svv%2BzVYczf//73uu2223TXXXcpLi5OHTt2VOPGjd3b3nzzzdq7d6/69%2B%2BvrKwss78UAAAg/0zaJOnFF19UVlaWevXqpbvuuku33HKLRo4cKUnau3ev%2Bz5sffr00YMPPqipU6fq6quv1pdffqlly5a5g6HJkyerY8eOio%2BP18CBAxUSEqKnn37aeIFnCXCZfdacDxQWFqpGjRruz/v3768JEyZo6NChl66IKr4Lsinsdik8vOwqVwtciJAX6P83VLDZpJAQqaDA/y9ECAu3yLmcnTqV3Z6kc2e/vxAhO8vPrxhX2b3kIiNDlJNTYIkLEaLDa1S8kq8FBEg1akiFhda4EOF/F9xdambdbcRHX45P%2BGXSZsTXX3%2Btrl27aseOHSopKdG7776r7Oxs9ejRw9elAQBQbflr0mYlfnf1qFFdu3ZVYmKipk6dqmPHjqlRo0ZatGiRx209AADApVXdGiwz/OaaNkkaM2aMxowZ4%2BsyAAAAqsxvsmkDAAD%2BhaTNOJo2AABgOpo2435zFyIAAAD8FpG0AQAA05G0GUfSBgAAYAEkbQAAwHQkbcbRtAEAANPRtBnH4VEAAAALIGkDAACmI2kzjqQNAADAAkjaAACA6UjajKNpAwAApqNpM47DowAAABZA0gYAAExH0mYcSRsAAIAFkLQBAADTkbQZR9MGAABMR9NmHIdHAQAALICkDQAAmI6kzTiSNgAAAAsgaQMAAKYjaTOOpg0AAJiOps04Do8CAABYAEkbAAAwHUmbcSRtAAAAFkDSBgAATEfSZhxNGwAAMB1Nm3EcHgUAALAAkjYAAGA6kjbjSNoAAAAsgKQNAACYjqTNOJo2k2QX%2B//UBsqmSEk5JXYVFwf4upyKFef5uoIKBQbaFBISosLCAhUX%2B/dPqFNZub4uwSuBgf97nW76zO/nFFUvO7fQ1yVUKDDQpsgaNZRTUOT3r9Ho6DCfPTdNm3EcHgUAALAA/4%2BDAACA5ZG0GUfSBgAAYAEkbQAAwHQkbcbRtAEAANPRtBnH4VEAAAALIGkDAACmI2kzjqQNAACgHE6nU1OnTlXPnj3Vu3dvPfbYYzp16tR51//ggw/0%2B9//Xp06dVJcXJzefvtt97KXXnpJrVu3Vvv27T0eR48e9boekjYAAGA6KyZts2bNUmFhoVJSUlRUVKQpU6YoKSlJjz/%2B%2BDnr7tixQ9OnT9cLL7ygfv366YsvvtDEiRN15ZVXqkuXLpKk%2BPh4/fGPf7zoemjaAACA6cxq2rKyspSdne0xFh0dLYfDYWi/R48eVWpqqt577z1FRUVJkiZMmKApU6bo4YcfVlBQkMf6TqdT999/vwYNGiRJ6tu3r2JjY7VlyxZ302YUTRsAALCs5ORkLV682GNs0qRJSkhIMLTfXbt2yW63q2XLlu6xtm3b6sSJE9qzZ4/HuCT16dNHffr0cX9eXFys7OxsxcTEuMe%2B/fZbDR8%2BXLt371aDBg30yCOPqHfv3l7XRNMGAABMZ1bSNmzYMA0YMMBjLDo62vB%2BnU6nQkNDFRDwy3tzR0RESJJycnIq3D4pKUm1a9fW4MGDJUn169dXo0aNNG3aNDkcDiUnJ%2BuBBx7QunXrdOWVV3pVE00bAACwLIfDcdGHQteuXasZM2aUuywxMVEul6vS%2B3S5XEpKSlJKSopWrVql4OBgSdLQoUM1dOhQ93pjxozR%2B%2B%2B/r3Xr1mnq1Kle7ZumDQAAmM4fL0SIj49XfHx8ucu%2B%2BOIL5efnq6SkRHa7XVJZ%2BiZJdevWLXeb0tJSPfLII9qxY4feeustNWrU6ILP37BhQ2VlZXldL7f8AAAApistNedhltatW8vlcikjI8M9lpaWpvDwcDVt2rTcbZ555hl999135TZsL7/8sjZv3uwxlpmZWWFjdzaaNgAAgF%2BJiopSXFycFi1apGPHjunw4cNasmSJbr/9dgUGlh2oHD16tNavXy9J%2Bs9//qN169Zp2bJlqlOnzjn7czqdmjNnjvbs2aPTp09r%2BfLl2rdvn2699Vava%2BLwKAAAMJ0/Hh6tyNy5c/Xkk09q4MCBCgoK0k033aTExET38v379%2Bv48eOSpNWrVysvL0/9%2B/f32EfXrl21fPlyTZs2TVLZuWxOp1PNmzfX66%2B/rvr163tdT4DrYs6yQ4Wys/N8XUKFAgNtiowMUU5OgYqLLfjd5IeY06rHnFYt5rPqWWlOo6PDfPbc//qXOfu94QZz9uuPODwKAABgARweBQAAprPi4VF/Q9IGAABgASRtAADAdCRtxtG0AQAA09G0GcfhUQAAAAsgaQMAAKYjaTOOpA0AAMACSNoAAIDpSNqMo2kDAACmo2kzjsOjAAAAFkDSBgAATEfSZhxJGwAAgAWQtAEAANORtBlH0wYAAExH02Ych0cBAAAsgKQNAACYjqTNOJI2AAAACyBpAwAApiNpM67aJm1fffWV%2BvTpo8GDB/u6FAAAfvNKS815VCfVtmlbuXKlfve73yklJcXXpQAAAFSo2h4ezc/PV8eOHWWzVdu%2BFQCAS6a6pWJmqJZN25133qktW7boP//5jz744AMNGDBAKSkpys/PV%2BPGjfXQQw/pmmuu8XWZAAAAbtWyafvb3/6mUaNGqWPHjmrZsqVee%2B01rVu3ThEREVqzZo0efvhhffLJJwoKCvJqf1lZWcrOzvYYCwysLYfDYUb5VcZut3l8hHHMadVjTqsW81n1mFPvkLQZVy2btrPl5uYqMDBQtWrVkt1u12233aZbb721UodNk5OTtXjxYo%2BxiRMnavLkyVVdrinCw2v5uoTfHOa06jGnVYv5rHrM6YXRtBlX7Zu2G2%2B8UWvXrlWfPn3Uq1cv9evXTzfeeGOlmrZhw4ZpwIABHmOBgbWVk1NQ1eVWKbvdpvDwWsrNPamSEr6bqgJzWvWY06rFfFY9K81pZGSIr0uAAdW%2BaatTp47efvttbd26VR999JFefPFFvfXWW3rjjTcUGOjd9DgcjnMOhWZn56m42L%2B/ec8oKSm1TK1WwZxWPea0ajGfVY85vTCSNuOq/QH406dP6%2BTJk%2BrcubOmTZumlJQU7d69WxkZGb4uDQAAwK3aJ21PP/20nE6nZs%2BercjISKWnp6u0tFSXXXaZr0sDAOA3g6TNuGrftE2bNk1PPvmk4uLiVFxcrCuuuEILFixQVFSUr0sDAOA3g6bNuGrbtP31r391/3vRokU%2BrAQAAKBi1bZpAwAAlw5Jm3HV/kIEAAAAKyBpAwAApiNpM46mDQAAmI6mzTgOjwIAAFgASRsAADAdSZtxJG0AAAAWQNIGAABMR9JmHE0bAAAwHU2bcRweBQAAsACaNgAAYLrSUnMeZnI6nZo6dap69uyp3r1767HHHtOpU6fKXffdd99Vq1at1L59e4/Hjh07/vf1l2rhwoUaOHCgunbtqnHjxmn//v2VqoemDQAAmM6KTdusWbN08uRJpaSkaPXq1crMzFRSUtJ51%2B/atavS0tI8Hh06dJAkvfHGG/rnP/%2BpZcuW6aOPPlKTJk00ceJEuVwur%2BuhaQMAAPiVo0ePKjU1VYmJiYqKilJMTIwmTJig1atXq6ioqNL7S05O1pgxY9SsWTOFhoYqMTFRmZmZ2r59u9f74EIEAABgOrNSsaysLGVnZ3uMRUdHy%2BFwGNrvrl27ZLfb1bJlS/dY27ZtdeLECe3Zs8dj/IxDhw7p7rvv1s6dOxUeHq7JkycrPj5ep06d0vfff682bdq41w0NDdUVV1yhtLQ0/e53v/OqJpo2AABgWcnJyVq8eLHH2KRJk5SQkGBov06nU6GhoQoICHCPRURESJJycnLOWT8qKkpNmjTRgw8%2BqObNm2vjxo2aMWOGHA6HrrzySrlcLvf2Z%2B%2BvvH2dD00bAAAwnVlJ27BhwzRgwACPsejoaK%2B2Xbt2rWbMmFHussTExEqdb9avXz/169fP/fmNN96ojRs36t1339X06dMlqVL7Kw9NGwAAMJ1ZTZvD4bjoQ6Hx8fGKj48vd9kXX3yh/Px8lZSUyG63SypL3ySpbt26Xu2/YcOG2rlzp%2BrUqSObzebe/gyn0%2Bn1viQuRAAAADhH69at5XK5lJGR4R5LS0tTeHi4mjZtes76b731ltavX%2B8xlpmZqUaNGik4OFgtWrRQenq6e1lubq727dvnvrrUGzRtAADAdFa75UdUVJTi4uK0aNEiHTt2TIcPH9aSJUt0%2B%2B23KzCw7EDl6NGj3Y1aYWGhnnrqKaWlpamoqEgpKSn69NNPNXz4cEnSiBEjtGrVKmVmZio/P19JSUlq3bq12rdv73VNHB4FAAAox9y5c/Xkk09q4MCBCgoK0k033aTExET38v379%2Bv48eOSpLvuuksFBQWaMmWKsrOzdfnll2vJkiVq166dJGn48OHKzs7WqFGjVFBQoG7dup1zAUVFAlxGz4pDubKz83xdQoUCA22KjAxRTk6Biot5U7iqwJxWPea0ajGfVc9KcxodHeaz537kEXP2O3%2B%2BOfv1RyRtAADAdLxhvHGc0wYAAGABJG0AAMB0JG3GkbQBAABYAEkbAAAwHUmbcTRtAADAdDRtxnF4FAAAwAJI2gAAgOlI2owjaQMAALAAkjYAAGA6kjbjaNoAAIDpaNqM4/AoAACABZC0AQAA05G0GUfSBgAAYAEkbSaJvvIyX5dQsY4dpc8/V%2BTNcdL27b6upkI5%2B/J8XUKF7PYzH0N8W4gXImuf9nUJ3gkIkCRFhgRJLpePi7mw7NxCX5fwmxPtCPd1CRXr1EnaulWRA6%2BRtm3zdTUV89H3EUmbcTRtAADAdDRtxnF4FAAAwAJI2gAAgOlI2owjaQMAALAAkjYAAGA6kjbjaNoAAIDpaNqM4/AoAACABZC0AQAA05G0GUfSBgAAYAEkbQAAwHQkbcbRtAEAANPRtBnH4VEAAAALIGkDAACmI2kzjqQNAADAAkjaAACA6UjajKNpAwAApqNpM47DowAAABZA0gYAAExH0mYcTRsAADAdTZtxHB4FAACwAJI2AABgOpI240jaAAAALICkDQAAmI6kzTiaNgAAYDqaNuM4PAoAAGABJG0AAMB0JG3GkbQBAABYAEkbAAAwnRWTNqfTqdmzZ%2Bvf//63bDab%2Bvbtq1mzZqlmzZrnrPv4449r7dq1HmMlJSWKj4/X/PnzNXPmTK1bt052u929PDg4WFu2bPG6HpI2AABgutJScx5mmjVrlk6ePKmUlBStXr1amZmZSkpKKnfdefPmKS0tzf3Ytm2brrzySl1//fXudcaPH%2B%2BxTmUaNommDQAA4BxHjx5VamqqEhMTFRUVpZiYGE2YMEGrV69WUVFRhduvXLlSl112mfr27VtlNXF4FAAAmM6sVCwrK0vZ2dkeY9HR0XI4HIb2u2vXLtntdrVs2dI91rZtW504cUJ79uzxGP%2B13Nxcvfrqq3rzzTc9xr/66itt2rRJP/74o5o1a6bZs2erXbt2XtdE0wYAACwrOTlZixcv9hibNGmSEhISDO3X6XQqNDRUAQEB7rGIiAhJUk5OzgW3/dvf/qauXbuqRYsW7rFGjRrJZrNpypQpCgkJ0eLFizV27Fht2LBBkZGRXtVE0wYAAExnVtI2bNgwDRgwwGMsOjraq23Xrl2rGTNmlLssMTFRLper0vWUlJTojTfe0IIFCzzGJ06c6PH5Qw89pJSUFKWmpmro0KFe7ZumDQAAmM6sps3hcFz0odD4%2BHjFx8eXu%2ByLL75Qfn6%2BSkpK3Fd8Op1OSVLdunXPu8%2Bvv/5ahYWF6tKlywWf2263q0GDBsrKyvK6Xi5EAAAA%2BJXWrVvL5XIpIyPDPZaWlqbw8HA1bdr0vNtt2rRJ3bt3V2DgL7mYy%2BXS/PnzPfZVWFioffv2qVGjRl7XRNMGAABMZ7VbfkRFRSkuLk6LFi3SsWPHdPjwYS1ZskS33367uyEbPXq01q9f77Hdrl27dPnll3uMBQQE6MCBA5ozZ46OHDmigoICJSUlKSgoSIMGDfK6Jpo2AACAcsydO1dhYWEaOHCgfv/736tDhw5KTEx0L9%2B/f7%2BOHz/usU12drbq1at3zr6efvppNWnSREOGDFHPnj21a9curVy5UrVr1/a6Hs5pAwAAprPiOyKEhYXphRdeOO/yDz/88JyxDRs2lLtunTp1NH/%2BfEP1VCppO3jwoNq3b6%2B9e/dW%2BomSkpI0atSoSm9ntpkzZ3p0zQAAoOpZ7fCoP6pU0tawYUOlpaWZVQsAAADOg8OjAADAdNUtFTNDpZq2AwcOaODAgVq/fr3uvfdejR8/Xhs3btTXX3%2BtunXravbs2erdu7eksuO8zz77rLKystS3b99yT8o7n5kzZyokJESBgYF67733ZLPZNG7cON17772SpJYtW%2Bq1115Tnz59JElvvfWWXnvtNX344YfuGpcuXapnn31WP/30kwYPHqzx48drxowZysjIUPv27bV48WL3nY1dLpfmzZunNWvWKDQ0VOPGjavUodxy30LjiivkqFPH6334RGys50c/97/b5Pg1m83zo1876y7ffu1MnRaoNzDQ///j7Xabx0e/16mTryuoWKtWnh/92bZtvq4ABhhK2v7yl7/oueeeU6tWrTR79mw988wzWr9%2BvXJzc5WYmKjp06dr2LBh2rx5s6ZNm6bWrVt7ve%2BUlBTNnDlTX3zxhdatW6dZs2YpPj7e6xvorVmzRm%2B//bYyMjJ055136scff9Tzzz%2BvmjVr6tZbb9Xq1as1duxYSdKnn36qhx56SF9%2B%2BaU%2B/vhjJSQkqGPHjurQoYNXz1XuW2hMmKCEKVO8/np9avlyX1fglXBfF1AJoaG%2BrsAbNXxdQOUEBfm6ggpF1rDOnIaH1/J1Cd7ZutXXFXjvV%2B8z6Zd8%2BMcPSZtxhpq2/v37uxubuLg4rVmzRqWlpfr8889Vu3Zt/eEPf5DNZlPfvn3VpUsXFRQUeL3vyy%2B/XLfeeqskafDgwXr00Uf1ww8/eN203XbbbQoLC1PXrl0VFhamXr16uW9g16FDB/3www/udR0Oh0aMGCFJuu6669S6dWt9%2BumnXjdt5b6FxrBh0j/%2B4dX2PhMbW9awjR0r7d7t62oqlLv%2Bc1%2BXUCGbraxhy8/3/x9Q4TULfV2CdwICyhq2oiLpIt5S5lLKKSjydQkVstttCg%2Bvpdzckyop8fMXqaTIgdf4uoSKtWpV1rCNHCmddfNUePL3n4lWYKhpO/vmcTVr1lRJSYmKiop0%2BPBhNWjQQLazjhE1adJE6enpF7XvWrXK/iI8deqU19s3aNDA/e/g4GDFxMR4fF5Y%2BMsvrObNm3ts27hxYx05csTr5yr3LTR%2B/LHsYXBDiVkAABR2SURBVAW7d0vbt/u6igqVlPi6Au%2BVllqgXj9vgM7hcvl9zcXF1vmtVFJSao16rXQ4LyPDWvXCcgw1bbbznLhTWFiokl/9xiqtZIt9vn2Xp7x9B/wqAr7Q/n69zOVyKTg42OvnBwAAF0bSZpwpZ6I6HA4dOXJErrP%2BKs7MzKyy/deoUcMjddu3b5%2Bh/f36vnP79u276DefBQAAMIMpTVvPnj2Vn5%2Bvv//97yosLFRqaqq2V%2BHhtyZNmig1NVXFxcVKS0vTxx9/bGh/Bw4c0HvvvaeioiJt2rRJ3377ra699tqqKRYAAHBz3SpgStNWv359LViwQMuXL9fVV1%2BtdevWaeTIkVW2/0cffVTbtm1Tly5d9Kc//cl9FejFiouL0/bt29W9e3c99dRTmjt3rpo2bVpF1QIAAJo24wJcLj8/s9eqwsJ8XUHFOnaUPv9c6t3bEhci5OzL83UJFbLbpfBwKTfX/y9EiKx92tcleCcgQKpRQyos9PsLEbJz/f%2BK3MBAmyIjQ5STU2CJCxGiHRa42U%2BnTmW3Junc2RoXIvjo%2B6htW3P2W4lrHC2Pd0QAAACmq26pmBkuedP21FNP6e233z7v8vHjx2vChAmXsCIAAAD/d8mbtlmzZmnWrFmX%2BmkBAIAPkbQZx%2BFRAABgOpo24yzyjsEAAADVG0kbAAAwHUmbcSRtAAAAFkDSBgAATEfSZhxNGwAAMB1Nm3EcHgWA/9/evQdHVd5/HP9syIRAKhQhiRBimhYHqNjcnI0QCyRtCGXABAdIoYW2qDOOiGO2RRtopXYw9EbQCVIVqiX1D4M2GkgdkLsgI5cRzUWC5SYOobnQBJZKsmGX3x%2BO0f0lQEiyPJw979fMzmyec2A/cwjwzfd5znMAwALotAEAgICj09ZzdNoAAAAsgE4bAAAIODptPUfRBgAAAo6ireeYHgUAALAAOm0AACDg6LT1HEUbAAAIOIq2nmN6FAAAwALotAEAgICj09ZzdNoAAAAsgE4bAAAIODptPUfRBgAAAo6ireeYHgUAALAAOm0AACDg6LT1HJ02AAAAC6DTBgAAAo5OW89RtAEAgICjaOs5pkcBAAAsgE4bAAAIODptPUenDQAAwAIo2gAAQMD5fIF5BVplZaUyMzM1a9asa55bXFysrKwsJScna/bs2aqqqmo/1traqqeeekrjx49XamqqHnvsMTU1NV1XFoo2AAAQcFYs2jZs2KCFCxcqLi7umudu375dRUVF%2BtOf/qS9e/cqPT1dDz/8sD7//HNJ0sqVK1VdXa2SkhJt3rxZly9fVn5%2B/nXloWgDAADoRGtrq0pKSpSQkHDNc0tKSnT//fcrISFB4eHhevDBByVJO3bs0KVLl/TGG2/okUce0dChQ/XNb35Tjz/%2BuHbu3Km6urou5%2BFGBAAAEHCB6orV19eroaHBbywyMlJRUVE9/r1nzpzZ5XOrq6s1ZcqU9q9DQkI0evRoVVZWavTo0XK73brzzjvbj3/nO99ReHi4qqurFR0d3aXPoGgLFLfbdIJrqq%2BvV0lRkXJLS3vlmzvQBpkO0AX19fVat65Eubm5FrimfU0H6JL6%2BnqVlFjjmkZG3vzXtL6%2BXkVFL1viekqSLl82neCa2v8t3bTJGtfUkED9URYVlWjVqlV%2BY48%2B%2BqgWLlwYmA%2B8gubmZg0cONBvbODAgWpqalJzc7MkacCAAX7HBwwYcF3r2pgetbGGhgatWrWqw08o6D6uae/jmvYurmfv45qalZubq9LSUr9Xbm5ul35tWVmZRo4c2emrtLT0urNcvkZleq3j10KnDQAAWFZUVFS3O5zZ2dnKzs7ulRyDBg1q76h9qbm5WXfccYduvfXW9q8jIiLaj587d06DBw/u8mfQaQMAAOihMWPGqLq6uv1rr9erjz/%2BWAkJCYqNjdXAgQP9jn/yySfyeDwaM2ZMlz%2BDog0AAKAbJk%2BerIMHD0qSZs%2BerbfeeksffvihLl68qL/%2B9a8KCwvTxIkT1adPH82aNUsvvPCCzpw5o6amJhUWFiozM1NDhgzp8ucxPWpjkZGRevTRRxUZGWk6StDgmvY%2Brmnv4nr2Pq5p8MrKylJtba28Xq98Pp/uuusuSdKmTZsUExOjEydOtO/DNn78eLlcLj3%2B%2BOM6e/as7rrrLr300ksKDw%2BXJD322GP63//%2Bp%2BzsbF26dEnp6en63e9%2Bd115HJd7uioOAAAAAcf0KAAAgAVQtAEAAFgARRsAAIAFULQBAABYAEUbAACABVC0AQAAWABFGwAAgAVQtAEAAFgARRsAAIAFULQBPTB79my99tpram5uNh0FABDkeIwV2nm9XvXp08d0DEt5/vnntWXLFh07dkxpaWm677779IMf/EB9%2B/Y1Hc3SfD6f/vvf/8rj8XQ4NmzYMAOJrO/rf799Pp9qamo0dOhQDRo0yHAyAF1F0WYzCxYs0NNPP60hQ4b4jVdUVOg3v/mNNmzYYCiZtX322WfasmWLtmzZoqNHj%2BqHP/yhpk2bpnHjxpmOZjlvv/22nn76aZ0/f95v/PLly3I4HDp8%2BLChZNa1Z88e/frXv9aePXt06dIl/fSnP9WRI0ckSYWFhUpPTzec0Lqqqqq0a9cu1dXVqW/fvoqOjlZmZqbi4uJMR0MQomizGZfLpT179mjRokWaOXOmWlpatHLlSq1fv17z58/XwoULTUe0NI/HozfeeEOFhYW6cOGCYmJi9NBDD%2BnHP/6x6WiWkZaWphkzZmjy5Mmddiy//e1vG0hlbTk5OZo3b57uv/9%2BlZWVqaioSBs3blRVVZWWL1%2Bu0tJS0xEt6dVXX9WyZcs0evRoxcbGSpJOnTqlTz75RL///e81Y8YMwwkRbCjabOjgwYMqKChQ//799Z///EcjRozQkiVL2v/RwfV7//33tXHjRr3zzjuKiIjQ1KlTlZOTo8bGRi1fvlxOp1NLliwxHdMSUlJStG/fPoWGhpqOEjSSkpL0wQcfyOFwKC8vT7fffrvy8vIkScnJyfrggw8MJ7SmjIwM5efnKzMz029827ZteuaZZ7R9%2B3ZDyRCsuBHBhhITE5WVlaXKyko1NTVp2rRpFGzd9Mc//lETJkzQI488ora2Nj333HPasWOHfvWrX2nEiBG65557tGbNGr311lumo1rG1KlTtX//ftMxgkpERITOnz%2BvlpYWvffee%2B3ToU1NTRTHPeB2uzVx4sQO4xMmTOgwvQ/0Bv622syuXbtUUFCg2267TRs3btSZM2e0dOlSrV%2B/XkuXLmXq6TodPnxYeXl5mjRpkvr379/pOVFRUXrooYducDJrKSwsbH/fv39/5efnKzk5WcOHD5fD4fA71%2BVy3eh4ljdt2jT97Gc/U58%2BfRQfH6/ExES1tLRo6dKlSktLMx3PsrKzs/Xmm29q1qxZfuPl5eW67777DKVCMGN61GZSU1O1aNEiv7UWHo9Hzz//vIqLi3Xo0CGD6WBXc%2BfO7dJ5DodDxcXFAU4THKZOnary8nJJUlZWlhYuXKjz589r6tSpGjBggDwej5YtW6ZFixbplltuMZzWOn75y1%2B2v/f5fNq9e7eioqI0YsQIORwOnThxQqdPn9akSZO0fPlyg0kRjCjabObs2bMaPHhwp8f%2B/e9/64477rjBiQB/X94l%2Bv95vV41NDTotttuM5DKeiZMmKCkpCTdfvvtevnllzV//vwrnkv3suvy8/O7fC5FG3obRZsNffjhhyorK1NdXZ1Wr14tn8%2Bnd955R5MnTzYdDVBCQoI%2B%2BuijDuNut1sZGRk6cOCAgVTWs2/fPq1bt04XLlzQgQMHdPfdd3d6Ht1LwDpY02Yz69ev1x/%2B8Af96Ec/0u7duyVJDQ0NKigoUH19vebNm2c4Iexq8%2BbN2rx5s9ra2vymoL5UW1vL5s/XITU1VampqZK%2BmH7%2Bxz/%2BYThR8HG73frnP/%2BpkydPqrW11e%2BYw%2BFQQUGBoWQIVtw9ajNr1qzRmjVr9Mwzz7RPQUVHR%2BvFF1/Uq6%2B%2Bajgd7Oy73/2u7rzzTklSWFhYh9fIkSO1atUqwymtiYItMBYsWKC//e1vamhoUGtrq9%2BrpaXFdDwEITptNtPY2Kjk5GRJ8ls3NGLECNXX15uKBSg2NlYPPPCAHA7HVddfATeLyspKbd269YrrhIHeRtFmM3FxcXr//fc1duxYv/Hy8nKe6Yibwueff37FjlpISIiio6M1duxYvl9h3MiRIztMiwKBxI0INvP2229ryZIlysjI0KZNmzR79mwdOXJEhw4d0ooVK5SVlWU6Imxu/vz5qq6uVktLi771rW8pJCREJ06cUL9%2B/RQbG6vGxkY1Njbqueee45mZMOqzzz5Tfn6%2B0tLSFB0drZAQ/xVHOTk5hpIhWFG02VBlZaXefPNNnTp1SuHh4YqNjdXMmTPZWBc3hXXr1unYsWN64okn9I1vfEOSdOHCBf35z39WYmKipk%2BfrtLSUv3973/Xhg0bDKeFnS1evFilpaW65ZZbOn1O7p49ewykQjCjaANwU0lLS9O2bdsUHh7uN97a2qrJkydrx44d8vl8SklJYTNoGJWYmKi1a9decTsVoLexps0G5s6d2%2BlmpZ1hvyaY1tbWpo8//rj9hpkvHTlyRM3NzZKkqqoq3XrrrSbiAe2GDRumUaNGmY4BG6Fos4HExMT29xcvXlRZWZlSUlIUHx8vn8%2Bno0ePqqKiQnPmzDGYEvjCz3/%2Bc/3iF7/Q97//fcXExCg0NFS1tbXatWuXZsyYIY/Ho3nz5ikvL890VNjck08%2BqaeeekozZszodE1bfHy8oWQIVkyP2ozL5VJOTo7Gjx/vN75161aVl5fr2WefNZQM%2BMqWLVv07rvvqqGhQT6fT4MHD1Zqaqqys7PlcDh08OBBpqRgXGddNofD0f4otsOHDxtIhWBG0WYzycnJ2r9/v0JD/ZusbW1tcjqdrBECgC46ffr0VY/HxMTcoCSwC6ZHbSYqKkolJSX6yU9%2B4jdeWlqqyMhIQ6mAr9TW1mrt2rU6duxYp3tgvfbaawZSAR1RlOFGo2izmSeeeEIul0urV6/W0KFD5fV6VVdXJ7fbrcLCQtPxALlcLl28eFH33nuv%2BvXrZzoOcEUZGRlXvclr27ZtNzAN7IDpURtyu93avXu36urq5PF4FBUVpXHjxik6Otp0NEBJSUnatWuXBgwYYDoKcFX/v%2Bvr9Xp16tQp7dy5Uw8%2B%2BKBmzpxpKBmCFUWbjXk8ng5jYWFhBpIAX5k%2BfbrWrFmjIUOGmI4CdEtlZaWKior00ksvmY6CIEPRZjN79%2B5VQUGBTp48Ka/X2%2BE4dzvBtL1796q4uFhz5sxRTEwM2yjAcrxer%2B6%2B%2B25u7EKvo2izmczMTDmdTk2aNKnT9UJOp9NAKuArbKMAq%2BjsMVUtLS3aunWrDh06pM2bNxtIhWBG0WYzSUlJOnDgQIctP4CbBdsowCo6%2BwEjLCxMcXFxWrx4scaOHWsgFYIZ/3PbjNPpVE1NjcaMGWM6CtCpL4uys2fP6syZM3yv4qZVU1NjOgJshk6bzbz%2B%2But65ZVXlJ6eruHDh3dYL5Sbm2soGfCFuro6LV68WO%2B9955CQ0NVVVWl%2Bvp6PfDAA1q9erViY2NNRwTaNTY26uTJk2ppaelw7N577zWQCMGMos1mMjIyrnjM4XCwrxCMW7Bggfr37y%2BXy6WsrCxVVFTo0qVLWrFihY4fP64XX3zRdERAkrR27VqtXLmy05u6WH%2BJQKBos4FPP/1UcXFxkqTjx49fdTNI7syDaSkpKXr33XcVERGhhIQEffTRR5K%2BWOA9YcIE7du3z3BC4Atjx47VokWLNGXKFIWHh5uOAxtgTZsNTJs2TRUVFZKkKVOmdFq0cWcebhb9%2BvVTZz9Lnjt3rtOOBmCKz%2BdTdna2%2BvTpYzoKbIKizQY2bdrU/p7pT9zs7rnnHi1evFh5eXmSpPPnz6umpkZ/%2BctfNHHiRLPhgK%2BZPn26ysvLlZ2dbToKbILpUQA3lXPnzunJJ5/Uzp0728dCQkI0ZcoU/fa3v9XAgQPNhQO%2BZtmyZfrXv/6l4cOHd3pj14oVKwwlQ7CiaANg3IkTJzqMnT17VmFhYe2PVuvbt68k1l3i5pGfn3/V48uXL79BSWAXFG0AjBs1alT7Uw8kse4SQeWFF17Qww8/bDoGggBFGwDjrvUUhK/jiQiwmq/fBQ30BDciADCOQgzBjN4IekvItU8BAADddbW9MYHrQdEGAABgARRtAAAAFkDRBgAAYAEUbQAABBA3IqC3ULQBABBAs2bNMh0BQYJ92gAA6KK5c%2Bd2%2BW7Q4uLiAKeB3bBPGwAAXZSYmNj%2B/uLFiyorK1NKSori4%2BPl8/l09OhRVVRUaM6cOQZTIljRaQMAoBtcLpdycnI0fvx4v/GtW7eqvLxczz77rKFkCFYUbQAAdENycrL279%2Bv0FD/Sau2tjY5nU4dOnTIUDIEK25EAACgG6KiolRSUtJhvLS0VJGRkQYSIdjRaQMAoBu2b98ul8uliIgIDR06VF6vV3V1dXK73SosLFRmZqbpiAgyFG0AAHST2%2B3W7t27VVdXJ4/Ho6ioKI0bN07R0dGmoyEIUbQBANBDHo%2Bnw1hYWJiBJAhmbPkBAEA37N27VwUFBTp58qS8Xm%2BH44cPHzaQCsGMThsAAN2QmZkpp9OpSZMmqV%2B/fh2OO51OA6kQzCjaAADohqSkJB04cKDDlh9AoLDlBwAA3eB0OlVTU2M6BmyEThsAAN3w%2Buuv65VXXlF6erqGDx%2BukBD/Pkhubq6hZAhWFG0AAHRDRkbGFY85HA5t27btBqaBHVC0AQDQRZ9%2B%2Bqni4uIkScePH5fD4bjiufHx8TcqFmyCog0AgC763ve%2Bp4qKCknSqFGjOi3aLl%2B%2BLIfDwZYf6HUUbQAAdFFtba2GDRsmSTp9%2BvRVz42JibkRkWAjFG0AAAAWwJYfAAAAFkDRBgAAYAEUbQAAABZA0QYAAGABFG0AAAAWQNEGAABgARRtAAAAFvB/QyYyYRZtWJQAAAAASUVORK5CYII%3D\" class=\"center-img\">\n",
       "    <img src=\"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAm0AAAIiCAYAAACaBf8HAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAAPYQAAD2EBqD%2BnaQAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvhp/UCwAAIABJREFUeJzs3XlclPXe//E3A4jKJiij5pIrrmiY5popVpRltGgudy7laVNROZrZLy31WJ4KlzuxRUvTU3noZEfNPLeJLZ7KOpmWSJKGlLtQMiIoss3vD3OOkyjoxdXMJa/n48ED%2BV4Ln/k6DB/e1zI%2BTqfTKQAAAHg1m6cLAAAAQPlo2gAAACyApg0AAMACaNoAAAAsgKYNAADAAmjaAAAALICmDQAAwAJo2gAAACyApg0AAMACaNoAAAAswM/TBQCoPDt27NB7772n7du368CBAzp16pQCAgJkt9vVvn179e/fXzExMfLx8fF0qQCAS%2BTDe48C1ud0OvXMM8/ob3/7myQpNDRUkZGRCgoKUl5ennbt2qW8vDxJUt%2B%2BffXiiy%2BqWrVqniwZAHCJaNqAK8Bbb72lWbNmqXr16po5c6YGDBggX19f1/KCggL9/e9/13PPPafS0lI99NBDmjRpkgcrBgBcKs5pA64AK1eulCSNGDFCd955p1vDJknVq1fXqFGj9OCDD0qS3nzzTRUUFPzhdQIALh9NG3AF2LdvnySpVatWF11v1KhReu2117Rx40ZVr15dkjR16lS1atVKzz33nPLz8zVnzhzFxMQoKipK3bp107hx47R79%2B4L7nPLli0aN26cevXqpfbt26tbt24aNWqU3n///Qtu43A4NH/%2BfN15552Kjo5Wu3bt1KNHDz3yyCP66quvytwmJiZGrVq10ubNm7Vx40b1799fUVFR%2BuabbyRJCxcuVKtWrTR16lQVFhZq4cKFio2NVYcOHXTDDTfoL3/5i06ePClJ2rNnj8aPH6%2BePXsqKipKAwYM0Nq1a8v8vqdOndKSJUt07733qnPnzmrbtq26du2q%2B%2B%2B/Xxs3bixzm%2BHDh6tVq1ZauXKl8vPzNX/%2BfFctXbp00YMPPqidO3decH4AoCw0bcAVIDw8XJL07bfflrve9ddfrzp16py3LD8/X6NGjdLKlSsVFhamLl26qKSkRBs3btSgQYPKbDLmz5%2BvUaNGaePGjQoKClLXrl0VHh6uLVu2aPLkyZo4caJKSkrctsnOztbdd9%2BtV155RXv37lXbtm3VrVs3VatWTR9//LFGjhypVatWXfAx7N27VwkJCfL391e3bt1czee5HnvsMS1dulQNGjRQy5YtlZWVpTfffFOPP/64du/eraFDhyo9PV2tW7dWRESEdu/erccee0yffPKJ235OnTqlYcOGKTExUWlpaWrevLm6d%2B%2Bu0NBQffHFFxo3bpwWLlx4wVoLCws1atQoLV%2B%2BXPXr11e7du1UUlKizZs3a/jw4a5mGwAqxAnA8h5//HFnZGSks1WrVs7nnnvOefTo0UvetlOnTs7bb7/dbdvc3FznwIEDnZGRkc6BAwe6bffhhx86IyMjnV26dHF%2B9dVXbsv%2B85//OHv06OGMjIx0Llu2zG3Z7NmznZGRkc7Y2FjnL7/84hovLi52PvPMM87IyEjntdde68zLy3Pbrm/fvs7IyEhnTEyM8%2BWXXz7vcbz44ovOyMhIZ69evZx33HGHMzs727Vs48aNrvm55ZZbnAsXLnSWlpY6nU6ns6SkxDl27FhnZGSkc/To0W77XLZsmTMyMtLZrVs3Z2ZmZpnL2rRp49y3b5/bsvvuu88ZGRnp7NOnj3PYsGHOX3/91bUsKyvL2bNnT2dkZKTz2WefPe9xAMCFkLQBV4CJEyeqfv36cjqdev3119WnTx/dd999WrBggb744gudOnWq3H3k5eVpxowZstvtrrHg4GBNnTpV0pnbifz000%2BuZS%2B//LIk6fHHH9d1113ntq8uXbpoypQpkqTly5e7LbPb7brttts0btw41a5d2zXu6%2BurhIQE2Ww2nThx4oKp4dkLKS4kKytLM2fOdEsTb7zxRkVERMjpdMrPz09jx4513fbEZrNp8ODBkqTvv//ebV%2BBgYEaMGCAHnnkETVp0sRt2ciRIxUREaGSkhJt2bKlzFqys7M1b948VxIqSRERERowYIAk6bvvvrvg4wCA3%2BM%2BbcAVoF69evrHP/6hF154QevWrVNJSYm%2B/vprff3113r55Zfl7%2B%2Bvrl276o477tBtt90mP7/zf/Ttdruuvfba88ajo6MVGBio/Px8ffvtt2rSpImOHj2qtLQ0SdJNN91UZk033nijfHx8dOjQIWVmZqpp06aS5LoYoiw1atRQ7dq1lZ2drezs7DLX6datm2y2C/%2B9GRERoWuuuea88QYNGig7O1t9%2B/Y97z51DRo0kCQdP37cbXzQoEEaNGhQmd/Hx8dHDRs2vGit1113nerWrXveeOPGjSVJOTk5F3wcAPB7NG3AFSIiIkLPP/%2B8pkyZoo0bN2rLli3aunWrfv31VxUVFemzzz7TZ599piVLluill15yNQ5nRUZGlrlfm82mBg0aaPfu3Tpw4IAkuV2YcDaJK4ufn5%2BKior0008/uZo26cy5Xp9%2B%2Bql27typrKws5ebmyvnb3YdOnDgh6UyiVpaymqBzNWzYsMzxs/elO9uglbWsuLj4vGWlpaX6/PPPtX37dmVlZcnhcLhqy8zMvGitjRo1KnM8ICBAklRUVHSxhwIAbmjagCtMnTp1NHToUA0dOlTSmRP3v/rqK61Zs0bbt2/Xnj179NBDD%2Bn999%2BXv7%2B/a7vQ0NAL7jM4OFjSmYsVJPdEatOmTeXWdLYRk6T09HSNHTvW1QBeqovVKcntMV3O8nMdOnRIY8eOPe%2BwaUVxA2MAlYmmDbjCNWvWTM2aNdPQoUO1fPlyPfvss8rMzNT69esVFxfnWu9ihxzPJkln1zl7eNHf31%2BpqakVflusgoICPfroozp06JCaNGmiMWPGqHv37goLC3M1UzExMTp48OAF9/FHvgXX%2BPHj9f3336tOnTqKj49X7969VadOHVczNnz4cP3nP//5w%2BoBULXRtAFVyMiRI/Xmm29q37592rNnj9uys29zVZazy84mbrVq1ZJ05vBeTk6O24n2F7N582YdOnRIPj4%2BWrx4sa6%2B%2Burz1vGWm/6mpaUpNTVVkpSYmKju3buft4631AqgauDqUcDiNm7cqKeeekqLFy%2Bu0PpnDy%2BePa/qrB9//LHM9UtKSlzJ19lztM49/%2B33zd/FnL36tHHjxmU2bPv27dOvv/5a4f2Z6eeff5Yk1/3gfi8/P/%2BiNx0GgMpG0wZYXGpqqpKTk7V48WIdOnToousePXpU6enpknTeFZb79%2B/Xrl27ztvmm2%2B%2Bcb2TQHR0tKQzFz20a9dOkvTOO%2B%2BU%2Bb0yMzMVFxfnujWI9N%2Bk7vTp02Vus2jRIte/f39T3j9aUFCQpDOHhsu6YOD11193JW1lXcAAAJWNpg2wuPvvv18RERE6ceKERo4cWebbQDmdTn355Zf605/%2BpKKiIl1zzTXq2bOn2zrBwcGaNm2aW9LlcDj017/%2BVdKZW21cddVVrmWPPPKIJGndunV64403XFd/SmdSqvHjxys9PV2HDx92jbdu3VqSdOTIEaWkpLjGT506pdmzZys1NdXVGO7du/ey56QyREZGytfXVyUlJXr77bdd48XFxXr11VeVnJysvn37SvrvVaQAYCbOaQMsLiwsTEuXLlV8fLx%2B%2BuknjRgxQhEREWrWrJlq1qwph8OhAwcOuO4ldu211yopKem8Cw9iYmK0f/9%2B9evXT1FRUfL399d3332nvLw8BQUF6amnnnJb/%2Babb9ZDDz2kxYsXa86cOVqxYoWaNWumnJwc7dq1SyUlJWrbtq0mT57s2iY6Olq9evXSZ599pvj4eHXs2FEBAQHauXOn/Pz8tGzZMn3wwQfavn27VqxYoR9%2B%2BEGPPvqounTpYv5E/k69evU0cOBAJScna86cOVq7dq1q1aqlXbt2KT8/X0lJSTpw4IA%2B/vhjbdiwQcOHD9ewYcN06623/uG1AqgaaNqAK0BkZKTef/99ffDBB/roo4%2BUnp6uHTt2qLCwUNWrV1fdunXVvXt39e/fX3369CnzCkybzaalS5fqpZde0ocffqhDhw4pMDBQsbGxSkhIcLvP2lmTJk1Sjx499Oabb%2Brbb7/Vli1bFBAQoPbt26t///4aNmzYebe9WLBggRITE7Vp0ybt3LlTdrtdsbGxeuSRR9S4cWPVq1dP33//vbZt26Y9e/b8oVeL/t60adMUGhqqdevWaffu3apdu7a6du2qhx9%2BWG3atFFBQYG%2B%2Buorbd68WXv27HFLGwGgsvk4eZUBqrSpU6fqn//8p%2B666y7XoVAAgPfhnDYAAAALoGkDAACwAJo2AAAAC6BpAwAAVda///1v9ejRQwkJCRddr7S0VPPnz1e/fv3UpUsXjR49Wvv373ctdzgcmjhxonr06KFevXrpySefrPR3TeFCBAAAUCUtWbJE7777rsLDw1WvXj3Nnz//guv%2B7W9/07Jly7RkyRLVrVtX8%2BfP19dff601a9bIx8dH8fHxKiws1Jw5c1RUVKQJEyaoffv2mjZtWqXVS9IGAACqpICAAL377rtlvq3e7yUnJ2vUqFFq3ry5goKClJCQoIyMDH333Xf65ZdflJKSooSEBIWHh6tu3boaM2aMVq1aVeY7qlwu7tMGAAAsKysry3Xz8LMiIiJkt9vL3XbEiBEV%2Bh4FBQX68ccf1bZtW9dYUFCQrr76aqWmpurEiRPy9fVVq1atXMvbtWunkydPau/evW7jRtC0mcWDNwStsKZNpT17pJYtJQu8Dc%2BJXO8/ku/jIwUGSvn5krefeBBckF3%2BSt7AZpPCw6Vjx6TSUk9Xc1HZqu7pEspls/koPDxQx47lq7TUy5%2BkkiLsIZ4uoXwWey312IuTSb8Xk198UUlJSW5j48aNU3x8fKV9j%2BPHj8vpdCo0NNRtPDQ0VDk5OapVq5aCgoLcbgZ%2Bdt2cnJxKq4OmrSqrVUvy9T3zGZXCx%2Be/H97etFmGzXZmQm02r2/arMBm85GPj49sNh9LNG2WwGupRw0ePFgxMTFuYxEREaZ8r4tdBvBHXCJA0wYAAMxnM%2Bc0ervdXqFDoUbUqlVLNptNDofDbdzhcKh27doKDw9XXl6eSkpK5Ovr61omSbVr1660OrgQAQAAmM9mM%2BfjDxAQEKCWLVsqLS3NNZabm6t9%2B/apQ4cOatOmjZxOp9LT013LU1NTFRISUub7Nl8umjYAAIDfOXr0qG655RbXvdiGDh2qFStWKCMjQ3l5eUpMTFSbNm0UFRWl8PBwxcbGasGCBTp27JiOHDmiRYsWaeDAgfLzq7yDmhweBQAA5vuDUrFLERUVJUkqLi6WJKWkpEg6k5IVFRUpMzNThYWFkqQhQ4YoOztbw4cPV35%2Bvrp27ep2AcSsWbP09NNPq1%2B/fvL399ftt99e7g17LxU31zWLFa4ejY6Wtm2TOnWStm/3dDXlssLVozbbf68e9fZz5i1z9aifnxQWJuXkSL%2B9sHorK1w96udnU1hYoHJy8lVc7OVPUlnk6lGLvZZ67CqpgABz9nv6tDn79UIkbQAAwHxemLRZDU0bAAAwH02bYcwgAACABZC0AQAA85G0GcYMAgAAWABJGwAAMB9Jm2E0bQAAwHw0bYYxgwAAABZA0gYAAMxH0mYYMwgAAGABJG0AAMB8JG2G0bQBAADz0bQZxgwCAABYAEkbAAAwH0mbYcwgAACABZC0AQAA85G0GUbTBgAAzEfTZhgzCAAAYAEkbQAAwHwkbYYxgwAAABZA0gYAAMxH0mYYTRsAADAfTZthzCAAAIAFkLQBAADzkbQZxgwCAABYAEkbAAAwH0mbYTRtAADAfDRthjGDAAAAFkDSBgAAzEfSZhgzCAAAYAFe3bQdPHhQUVFRyszMvORtExMTNXz4cBOqAgAAl8xmM%2BejCvHqw6MNGjRQamqqp8sAAABGVbEGywzMIAAAgAV4ddN24MABtWrVShkZGYqJidE//vEPPfTQQ4qOjtaNN96ozz77zLXuRx99pNjYWEVHR2vixIkqKChw29f69esVFxena665Rv369VNycrIk6ZdfftF1112nTz75xLXun//8Zz366KN/yGMEAKBK4PCoYV59ePT3Xn/9dT3//PNq3bq1ZsyYoWeffVbr169Xbm6uEhISNHnyZA0ePFhbtmzRpEmT1KZNG0lSamqqnnzySS1cuFDdu3fX9u3b9eCDD6ply5bq1KmTHnvsMc2ZM0c9evTQjh07tHnzZn3wwQcVrisrK0vZ2dluYxFNm8peq1alPv5K17q1%2B2cvZ4WfzbM1WqFW%2BVnkx9/X1/2zF/Pz7r%2BDJUm%2Bvja3z14vOtrTFZTPSq%2Bl27d7ugIYYJFX7TP69u2rDh06SJJiY2O1evVqlZaW6rPPPlPNmjX1P//zP7LZbLrhhhvUuXNn5efnS5Lee%2B899enTR7169ZIkde7cWbfeeqvWrFmjTp06aeDAgVq7dq2WLl2qDRs26LHHHlPdunUrXFdycrKSkpLcxsZNmKD4CRMq6ZGb7O23PV1BhQR6uoBLUKOGpyuogMAwT1dwaUJCPF1Buaw0oyEhVniSStq2zdMVVJwVXkt9fDz3vS3x16x3s1TT1rBhQ9e/q1evrpKSEhUVFenIkSOqX7%2B%2BbOc8IZo0aaK0tDRJ0r59%2B7RlyxZFRUW5ljudTlcT5%2BPjo9mzZ2vAgAFq37697r333kuqa/DgwYqJiXEbixgwQFq%2B/JIf4x%2BqdeszLzLDhknp6Z6uplz5//b%2BF2%2Bb7UzDduqUVFrq6WouLrAwx9MlVIyv75mGLTdXKinxdDUXlaNqni6hXL6%2BNoWE1FBu7imVlHj5k1RSWL/rPV1C%2BSz2WuoxNG2GWapps13gP7ywsFAlv3sxLz3nN2b16tU1dOhQTZ8%2B/YL7PnLkiPz8/HTkyBGdOnVKNWvWrHBddrtddrvdffAyblPiMenplojMvb0JOldpqQXqLS72dAWXpqTE62suttBLaklJqYqLvf1JKku8NrlY5LUU1nVFtL12u11Hjx6V0%2Bl0jWVkZLj%2B3bhxY/3www9u2xw5csTV6BUWFuqpp57SzJkz1bJlSy1YsOCPKRwAgKqCCxEMuyIebY8ePZSXl6e///3vKiwsVEpKir777jvX8oEDB2rbtm1atWqVCgsLtWvXLg0aNEgbNmyQJC1atEh2u10DBgzQk08%2BqXfeecdtewAAYBBNm2FXxKOtV6%2Be5s6dq6VLl%2Bq6667T2rVrNWzYMNfy5s2ba%2B7cuXrttdfUuXNnxcfHa/To0erfv792796t5cuX6%2Bmnn5Z0JpV74IEHNG3aNBUVFXnqIQEAALjxcZ57TBGVx5NX6FRUdPSZK7M6dbLEeRgncr3/qWqzSYGBUn6%2B95/TFlyQXf5K3sDPTwoLk3JyvP6ctmxV93QJ5fLzsyksLFA5OfmWOKctwu79Vw1b7bVUnvq137OnOfv9/HNz9uuFrHPWLAAAQCU7ePCgZs6cqe%2B%2B%2B041a9ZU//79NWnSpPMufnzggQf09ddfu40VFxdr7NixGjdunIYPH65t27a5bde0aVOtXbu20mqlaQMAAObz0vPP4uPj1a5dO6WkpOjXX3/Vww8/rDp16uj%2B%2B%2B93W2/p0qVuX%2Bfm5qp///666aabXGN/%2BctfdPfdd5tWq3fOIAAAuLJ44YUIqampSk9P1%2BTJkxUcHKwmTZpo1KhRrre6vJgFCxbopptuUqtWrQzVcClI2gAAgGWV%2BVaSERHn3z%2B1DGlpaWrQoIFCQ0NdY%2B3atVNmZqby8vIUFBRU5nY///yzVq9erZSUFLfx9evX67XXXtPhw4fVsWNHzZo1S40bN76MR1U2kjYAAGA%2Bk5K25ORk3X333W4fFUnKJMnhcCjkd2%2BRd7aBy8m58LvGLF68WPfcc4/Cw8NdY82bN1fLli319ttva9OmTQoPD9ef/vQnFRYWXsZklY2kDQAAWFaZbyUZEVHh7S/1JhoOh0Nr1qzRv/71L7fxGTNmuH09a9Ysde3aVd988426d%2B9%2BSd/jQmjaAACA%2BUy6EKHMt5KsoPDwcDkcDrcxh8MhHx8ftxTtXJs2bVLTpk3VqFGji%2B47KChIoaGhOnr06GXVVhYOjwIAAPN54YUI7du31%2BHDh3Xs2DHXWGpqqlq0aKHAwMAyt9m0aZN6/u6ec3l5eZoxY4Zbg3bs2DEdO3as3ObuUtC0AQCAKqlt27aKiorS3LlzlZeXp4yMDC1btkxDhw6VJN1yyy3aunWr2za7du1Sw4YN3caCgoL03Xffafbs2XI4HDp%2B/LhmzpypVq1aKTo6utLqpWkDAADm88KkTZJefPFFZWVlqWfPnhoxYoTuvPNO11thZmZm6uTJk27rZ2dnq06dOuftZ9GiRXI6nYqNjVWfPn1UVFSkxYsXn3eTXiM4pw0AAFRZ9erV05IlS8pc9sMPP5w3tnPnzjLXveqqq5SUlFSptf0eTRsAADCfl74jgpXQtAEAAPPRtBnGDAIAAFgASRsAADAfSZthzCAAAIAFkLQBAADzkbQZRtMGAADMR9NmGDMIAABgASRtAADAfCRthjGDAAAAFkDSBgAAzEfSZhhNGwAAMB9Nm2HMIAAAgAWQtAEAAPORtBnGDAIAAFgASRsAADAfSZthNG0AAMB8NG2GMYMAAAAWQNIGAADMR9JmGDMIAABgASRtAADAfCRthtG0AQAA89G0GcYMAgAAWABJGwAAMB9Jm2E0bSY5kev0dAnlstmkQEn5/96m0lJPV1O%2B4BAfT5dQvuhoads2BV7fSdq%2B3dPVXJQVnqPSOc/TamEq9fZXrIITnq7gipN3wvufpzabVFPSyc%2Bs8Voa5OkCcNm8/SUQAABcCUjaDKNpAwAA5qNpM4wZBAAAsACSNgAAYD6SNsNo2gAAgPlo2gxjBgEAACyApA0AAJiPpM0wZhAAAMACSNoAAID5SNoMo2kDAADmo2kzjBkEAACwAJI2AABgPpI2w5hBAAAACyBpAwAA5iNpM4ymDQAAmI%2BmzTBmEAAAwAJI2gAAgPlI2gxjBgEAQJV18OBBPfTQQ%2Bratav69u2rF154QaWlpeett3DhQrVp00ZRUVFuH7/88osk6fTp03rqqafUu3dvde3aVePHj1dOTk6l1krTBgAAzGezmfNhUHx8vOrWrauUlBQtW7ZMKSkpWr58eZnrxsXFKTU11e2jTp06kqT58%2BcrLS1NycnJ2rBhg5xOp5544gnD9Z2Lpg0AAJjPC5u21NRUpaena/LkyQoODlaTJk00atQoJScnX9J%2BiouL9e6772rMmDGqX7%2B%2BatWqpYkTJ%2BqTTz7R0aNHDdV4Ls5pAwAAlpWVlaXs7Gy3sYiICNnt9nK3TUtLU4MGDRQaGuoaa9eunTIzM5WXl6egoCC39X/44QcNGTJEu3fvVv369fXEE0%2BoV69e2rdvn06cOKF27dq51m3evLmqV6%2ButLQ01a1b1%2BCjPIOmDQAAmM%2BkCxGSk5OVlJTkNjZu3DjFx8eXu63D4VBISIjb2NkGLicnx61pq1evnho1aqRJkybJbrcrOTlZjzzyiNauXSuHwyFJ5%2B0rJCSkUs9ro2kDAACWNXjwYMXExLiNRUREVHh7p9NZofUGDRqkQYMGub4eNWqUPvjgA61du1a9e/e%2BpH1dLpo2AABgPpOSNrvdXqFDoWUJDw93pWRnORwO%2Bfj4KDw8vNztGzRooKysLNe6DodDgYGBruXHjx9X7dq1L6u2snAhAgAAMJ8XXojQvn17HT58WMeOHXONpaamqkWLFm7NlyS99NJL2rJli9tYRkaGGjVqpEaNGik0NFRpaWmuZbt371ZhYaHat29vqMZz0bQBAIAqqW3btoqKitLcuXOVl5enjIwMLVu2TEOHDpUk3XLLLdq6daukMynazJkztXfvXp0%2BfVpLly7Vvn37dNddd8nX11f33nuvXnnlFR0%2BfFg5OTmaN2%2BebrrpJtctQSoDh0cBAID5vPQdEV588UVNnz5dPXv2VFBQkIYMGaJhw4ZJkjIzM3Xy5ElJ0qRJkySdOZfN4XCoRYsWeuONN1SvXj1J0vjx45Wfn6%2B4uDgVFxerb9%2B%2BmjFjRqXW6uM0%2B6y5KurECU9XUD6bTQoMlPLzpTJu/ux1gkN8PF1C%2BaKjpW3bpE6dpO3bPV3NRZ3ItcaPvpWepwUF3v%2BD7%2BdnU1hYoHJy8lVc7OUTKqlGjWBPl1Aum02qWVM6edL7n6OS9Lu7WPxx5swxZ7%2BVfANbb0bSBgAAzOelSZuV0LQBAADz0bQZxgwCAABYAEkbAAAwH0mbYcwgAACABZC0AQAA85G0GUbTBgAAzEfTZhgzCAAAYAEkbQAAwHwkbYYxgwAAABZA0gYAAMxH0mYYTRsAADAfTZthVX4GR44cqb/%2B9a9uY4sWLdKQIUM8VBEAAMD5qnzSduedd2revHmaMmWKbL/9FfDhhx/q3nvvrfA%2BsrKylJ2d7TZWs2aE7HZ7pdZa2c7%2B0WOZP36ioz1dQflat3b/7MWs8v9upeepn5/3F%2Bnra3P77O2s8P/u4/Pfz95eb2mpB7%2B5t0%2BOBVT5pu3mm2/WrFmz9NVXX6l79%2B7av3%2B/MjIydOutt1Z4H8nJyUpKSnIbGzt2nMaPj6/sck1Ro4anK6igbds8XUHFvf22pysoV6CnC7hEVnieBgZaZ1ZDQiwwoRZjhedoXp6nK4ARVb5pCwwM1I033qi1a9eqe/fu%2BvDDD9WzZ0%2BFh4dXeB%2BDBw9WTEyM21jNmhHKz6/saiuXzXbmRebUKQ//9VVBgdd38nQJ5Wvd%2BkzDNmyYlJ7u6WouKv/f1miCrfQ8LSz08h96nUnYQkJqKDf3lEpKvHxCJQUEeH8j7OPz3%2Beo0%2BnparwYSZthVb5pk84cIh0/frxmzpypjRs36r777ruk7e12%2B3mHQk%2Bc8P5fMGeVllqk1u3bPV1BxaWne329lvg/P4cVnqfFxV5e4DlKSkotUa%2B/v6crKN/ZXsTp9P7nqEfRtBnGDErq3r27AgMD9Y9//EN79uwKb4VUAAAgAElEQVRRv379PF0SAACAG5I2STabTQMGDNC8efPUr18/1bDCiQkAAFgJSZthzOBv7rzzTuXl5WnAgAGeLgUAAOA8JG2/%2BfXXX9WgQQP17NnT06UAAHDlIWkzjKZNZ%2B6z9uyzz2r06NGue7UBAIBKxO9Xw6r8DL766qu69dZb1aVLFw0dOtTT5QAAAJSpyidtDz/8sB5%2B%2BGFPlwEAwJWNpM2wKt%2B0AQCAPwBNm2HMIAAAgAWQtAEAAPORtBnGDAIAAFgASRsAADAfSZthNG0AAMB8NG2GMYMAAAAWQNIGAADMR9JmGDMIAABgASRtAADAfCRthtG0AQAA89G0GcYMAgAAWABJGwAAMB9Jm2HMIAAAgAWQtAEAAPORtBlG0wYAAMxH02YYMwgAAGABJG0AAMB8JG2GMYMAAKDKOnjwoB566CF17dpVffv21QsvvKDS0tIy1125cqViY2MVHR2tuLg4paSkuJZNnTpVbdu2VVRUlOujc%2BfOlVorSRsAADCflyZt8fHxateunVJSUvTrr7/q4YcfVp06dXT//fe7rbdhwwbNnTtXr776qjp06KDVq1dr4sSJ%2Bte//qVGjRpJkh599FHFx8ebVqt3ziAAALiy2GymfGRlZSktLc3tIysrq0IlpaamKj09XZMnT1ZwcLCaNGmiUaNGKTk5%2Bbx1CwoK9Oc//1nXXnut/P39NWjQIAUGBurbb7%2Bt7Jm6IJI2AABgWcnJyUpKSnIbGzduXIUSr7S0NDVo0EChoaGusXbt2ikzM1N5eXkKCgpyjcfFxbltm5ubq/z8fNWtW9c19uWXX2rTpk36%2Beef1bx5c82YMUPt27e/3Id2Hpo2AABgPpMOjw4ePFgxMTFuYxERERXa1uFwKCQkxG3sbAOXk5Pj1rSdy%2Bl0atq0aerYsaOuu%2B46SVKjRo1ks9k0YcIEBQYGKikpSQ888IA2bNigsLCwS31YZaJpAwAAlmW322W32y97e6fTeUnrFxUVaerUqfrxxx%2B1YsUK1/jYsWPd1nvssce0bt06paSkaNCgQZdd37lo2gAAgPm88EKE8PBwORwOtzGHwyEfHx%2BFh4eft35BQYHGjBmjU6dO6a233rpogubr66v69etX%2BPy6ivC%2BGQQAAFceky5EMKJ9%2B/Y6fPiwjh075hpLTU1VixYtFBgY6Lau0%2BlUQkKC/Pz89MYbb7g1bE6nU3PmzFF6erprrLCwUPv27XNdWVoZaNoAAECVdPa%2BanPnzlVeXp4yMjK0bNkyDR06VJJ0yy23aOvWrZKk999/Xz/%2B%2BKP%2B93//VwEBAW778fHx0YEDBzRz5kwdPXpU%2Bfn5SkxMlL%2B/v2688cZKq5fDowAAwHxeeHhUkl588UVNnz5dPXv2VFBQkIYMGaJhw4ZJkjIzM3Xy5ElJ0qpVq3Tw4EHXhQdnxcXFafbs2XrmmWf03HPP6e6771ZeXp46dOig5cuXq2bNmpVWq4/zUs/AQ4WcOOHpCspns0mBgVJ%2BvnSBmz97leAQH0%2BXUL7oaGnbNqlTJ2n7dk9Xc1Encq3xo2%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%2BvbtqxdeeEGlpaVlrrtixQrFxsaqU6dOGjp0qHbu3Oladvr0aT311FPq3bu3unbtqvHjxysnJ6dSa61yTduHH36oDRs2eLoMAACqFi9N2uLj41W3bl2lpKRo2bJlSklJ0fLly89b76OPPtLChQv1/PPP64svvlDfvn31yCOP6OTJk5Kk%2BfPnKy0tTcnJydqwYYOcTqeeeOIJw/WdyysvRHA4HJo9e7a%2B/PJL5efnq1u3bpoxY4bq1q2rAwcOqF%2B/flq6dKleeOEF/fTTT2rdurUSExPVsGFDSdJLL72kZcuWyc/PTw8//LA%2B/fRTderUSTVr1nQdCk1JSdG3334rSSopKdFTTz2ldevWKSAgQNOnT1f//v0rXG9WVpays7PdxiL8/GS32ytpRkzi6%2Bv%2B2ctZIQU/W6MValV0tKcrqJjWrd0/ezE/P%2B//j/f1tbl99no%2BPp6uoHxna7RCrU6n5763SS%2BMZf4Ojoio0O/g1NRUpaena9myZQoODlZwcLBGjRql5cuX6/7773dbNzk5WXfffbc6duwoSfrTn/6kFStW6OOPP1ZsbKzeffddPffcc6pfv74kaeLEibrtttt09OhR1a1bt1Ieq1c2bVOnTpWfn58%2B%2BOAD%2Bfr66umnn9YTTzyhpUuXutZZsWKFXn31VQUEBGjEiBF67bXXNGPGDG3cuFGvvPKKli5dqrZt22rOnDlKS0tTp06dNHr0aO3Zs0enT5/W/PnzXftat26dnn32WU2bNk1JSUmaMWOGbr75Zvn5VWx6kpOTlZSU5DY2buxYxY8fXzkTYraQEE9XUCGBni7gEtSo4ekKKmDbNk9XcGneftvTFZQrzNMFXIKQECs8SS3G39/TFZTv9GlPV1DpyvwdPG6c4uPjy902LS1NDRo0UGhoqGusXbt2yszMVF5enoKCgtzWPTfQsdlsatOmjVJTU9WmTRudOHFC7dq1cy1v3ry5qlevrrS0tCu3aTt27Jg%2B/vhjrV%2B/3jWJkydPVp8%2Bfdw66aFDh7omoVevXkpNTZUkffrpp%2BrVq5c6d%2B4sSZoyZYrWrFlz0e/ZqVMnXX/99ZKkW265Ra%2B%2B%2BqqOHTtW4aRs8ODBiomJcRuL8POTKvlYdqXz9T3TsOXmSiUlnq6mXPnVvP9Xos12pmE7dUq6wCkRXiPw%2Bk6eLqFiWrc%2B07ANGyalp3u6movK2fRvT5dQLl9fm0JCaig395RKSrz8SSopLNACjZCPz5mGrajIs0mWl3PKnCSyzN/BEREV2tbhcCjkd8HF2d4jJyfHrWlzOBxuzd3ZdXNycuRwOCTpvH2FhIRU6nltXte0%2Bf/2l8qdd97pNu7r66vDhw8rPDxcklyHQiWpRo0aOv3bXw/Z2dlq3Lixa1lwcLCaNGly0e957r4CAgIkSYWFhRWu2W63n9/gZWdLxcUV3odHlZRYotZSr3u2Xlhpqfc3bdq%2B3dMVXJr0dK%2BvubjY2//T/6ukpNQa9VqpCXI6rVXvFaLM38GXwHkJ/2flrXsp%2B7ocXvdrsHr16pKkzZs3Kyzs/GTlwIEDkiSfC5w7UFpaet5hTVs5x9EvtC8AAFA5zPpD1shp2eHh4a6U7CyHwyEfHx9XSHRWWFhYmeu2bNnSta7D4VBg4H9P5jl%2B/Lhq1659%2BQX%2BjleeiWqz2fTDDz%2B4vi4qKtLRo0crtG3t2rV16NAh19d5eXnKzMys9BoBAEDFnT0CUdkfRrRv316HDx/WsWPHXGOpqalq0aKFW/N1dt20tDTX1yUlJfr%2B%2B%2B/VsWNHNWrUSKGhoW7Ld%2B/ercLCQrVv395YkefwuqbN399f/fv3V2Jioo4cOaKCggLNmzdPDzzwQIVix27dumnz5s3asWOHCgoK9Pzzz7vSO%2BnM4c/Dhw8rNzdXxRY4JAgAAMzRtm1bRUVFae7cucrLy1NGRoaWLVumoUOHSjpznvvWrVslnTmXfvXq1fr222916tQpvfzyy6pWrZr69OkjX19f3XvvvXrllVd0%2BPBh5eTkaN68ebrppptUp06dSqvX65o2SZo%2Bfbquvvpq3Xbbbbr%2B%2Buv1448/6qWXXqrQYcw77rhD99xzj0aMGKHY2Fh17NhRjRs3dm07YMAAZWZmqm/fvsrKyjL7oQAAAHln0iZJL774orKystSzZ0%2BNGDFCd955p4YNGyZJyszMdN2HrXfv3vrzn/%2BsiRMn6rrrrtMXX3yhxYsXu4Kh8ePHq2PHjoqLi1O/fv0UGBioZ555xniB5/Bxmn3WnAcUFhaqWrVqrq/79u2rMWPGaNCgQX9cEb%2B7Z4xX8vOTwsLOXOVqgdTxRPWKXQ3kSTabFBgo5ed7/4UIwSEWOZczOvrM7Uk6dfL6CxGys3I9XUK5/PxsCgsLVE5OviUuRIgIqVb%2BSp7m4yNVqyYVFlrjQoTfLrj7o5l1txEPPRyP8MqkzYivv/5aXbp00Y4dO1RSUqL33ntP2dnZ6t69u6dLAwCgyvLWpM1KvO7qUaO6dOmihIQETZw4UceOHVOjRo20YMECt9t6AACAP1ZVa7DMcMU1bZI0atQojRo1ytNlAAAAVJorsmkDAADehaTNOJo2AABgOpo24664CxEAAACuRCRtAADAdCRtxpG0AQAAWABJGwAAMB1Jm3E0bQAAwHQ0bcZxeBQAAMACSNoAAIDpSNqMI2kDAACwAJI2AABgOpI242jaAACA6WjajOPwKAAAgAWQtAEAANORtBlH0gYAAGABJG0AAMB0JG3G0bQBAADT0bQZx%2BFRAAAACyBpAwAApiNpM46kDQAAwAJI2gAAgOlI2oyjaQMAAKajaTOOw6MAAAAWQNIGAABMR9JmHEkbAACABZC0AQAA05G0GUfTBgAATEfTZhyHRwEAACyApA0AAJiOpM04kjYAAAALIGkDAACmI2kzjqbNJNmq7ukSyuUnm8Ik5aiaiq3wVCg44ekKyuXnZ1NgYKAKC/NVXOzdr1AFWbmeLqFC/Px%2Be55u%2BrfXzykqX3ZuoadLKJefn01h1aopJ7/I65%2BjERHBHvveNG3GcXgUAADAAiwQrwAAAKsjaTOOpA0AAMACSNoAAIDpSNqMo2kDAACmo2kzjsOjAAAAFkDSBgAATEfSZhxJGwAAQBkcDocmTpyoHj16qFevXnryySdVUFBwwfU//PBD3XHHHYqOjlZsbKzeeecd17KFCxeqTZs2ioqKcvv45ZdfKlwPSRsAADCdFZO26dOnq7CwUOvWrVNRUZEmTJigxMRETZs27bx1d%2BzYocmTJ2vevHnq06ePPv/8c40dO1bNmjVT586dJUlxcXH661//etn10LQBAADTmdW0ZWVlKTs7220sIiJCdrvd0H5/%2BeUXpaSk6J///KfCw8MlSWPGjNGECRP0%2BOOPy9/f3219h8Ohhx9%2BWDfeeKMk6YYbblBkZKS2bt3qatqMomkDAACWlZycrKSkJLexcePGKT4%2B3tB%2Bd%2B3aJV9fX7Vq1co11q5dO508eVJ79%2B51G5ek3r17q3fv3q6vi4uLlZ2drbp167rGfvjhBw0ZMkS7d%2B9W/fr19cQTT6hXr14VrommDQAAmM6spG3w4MGKiYlxG4uIiDC8X4fDoaCgIPn4%2BLjGQkNDJUk5OTnlbp%2BYmKiaNWuqf//%2BkqR69eqpUaNGmjRpkux2u5KTk/XII49o7dq1atasWYVqomkDAACWZbfbL/tQ6Jo1azRlypQylyUkJMjpdF7yPp1OpxITE7Vu3TqtWLFCAQEBkqRBgwZp0KBBrvVGjRqlDz74QGvXrtXEiRMrtG%2BaNgAAYDpvvBAhLi5OcXFxZS77/PPPlZeXp5KSEvn6%2Bko6k75JUu3atcvcprS0VE888YR27NihlStXqlGjRhf9/g0aNFBWVlaF6%2BWWHwAAwHSlpeZ8mKVNmzZyOp1KT093jaWmpiokJERNmzYtc5tnn31We/bsKbNhe%2Bmll7Rlyxa3sYyMjHIbu3PRtAEAAPxOeHi4YmNjtWDBAh07dkxHjhzRokWLNHDgQPn5nTlQOXLkSK1fv16S9M0332jt2rVavHixatWqdd7%2BHA6HZs6cqb179%2Br06dNaunSp9u3bp7vuuqvCNXF4FAAAmM4bD4%2BWZ9asWXr66afVr18/%2Bfv76/bbb1dCQoJr%2Bf79%2B3X8%2BHFJ0qpVq3TixAn17dvXbR9dunTR0qVLNWnSJElnzmVzOBxq0aKF3njjDdWrV6/C9fg4L%2BcsO5QrO/uEp0sol5%2BfTWFhgcrJyVdxsQV/mrwQc1r5mNPKxXxWPivNaUREsMe%2B97/%2BZc5%2Bb73VnP16Iw6PAgAAWACHRwEAgOmseHjU25C0AQAAWABJGwAAMB1Jm3E0bQAAwHQ0bcZxeBQAAMACSNoAAIDpSNqMI2kDAACwAJI2AABgOpI242jaAACA6WjajOPwKAAAgAWQtAEAANORtBlH0gYAAGABJG0AAMB0JG3G0bQBAADT0bQZx%2BFRAAAACyBpAwAApiNpM46kDQAAwAJI2gAAgOlI2oyrsknbl19%2Bqd69e6t///6eLgUAgCteaak5H1VJlW3ali9frmuuuUbr1q3zdCkAAADlqrKHR/Py8tSxY0fZbFW2bwUA4A9T1VIxM1TJpu2%2B%2B%2B7T1q1b9c033%2BjDDz9UTEyM1q1bp7y8PDVu3FiPPfaYrr/%2Bek%2BXCQAA4FIlm7Y333xTw4cPV8eOHdWqVSstWbJEa9euVWhoqFavXq3HH39cn376qfz9/Su0v6ysLGVnZ7uN%2BfnVlN1uN6P8SuPra3P7DOOY08rHnFYu5rPyMacVQ9JmXJVs2s6Vm5srPz8/1ahRQ76%2Bvrrnnnt01113XdJh0%2BTkZCUlJbmNjR07VuPHj6/sck0RElLD0yVccZjTysecVi7ms/IxpxdH02ZclW/abrvtNq1Zs0a9e/dWz5491adPH912222X1LQNHjxYMTExbmN%2BfjWVk5Nf2eVWKl9fm0JCaig395RKSvhpqgzMaeVjTisX81n5rDSnYWGBni4BBlT5pq1WrVp65513tG3bNn388cd68cUXtXLlSr311lvy86vY9Njt9vMOhWZnn1BxsXf/8J5VUlJqmVqtgjmtfMxp5WI%2BKx9zenEkbcZV%2BQPwp0%2Bf1qlTp9SpUydNmjRJ69at0%2B7du5Wenu7p0gAAAFyqfNL2zDPPyOFwaMaMGQoLC1NaWppKS0t11VVXebo0AACuGCRtxlX5pm3SpEl6%2BumnFRsbq%2BLiYl199dWaO3euwsPDPV0aAABXDJo246ps0/a3v/3N9e8FCxZ4sBIAAIDyVdmmDQAA/HFI2oyr8hciAAAAWAFJGwAAMB1Jm3E0bQAAwHQ0bcZxeBQAAMACSNoAAIDpSNqMI2kDAACwAJI2AABgOpI242jaAACA6WjajOPwKAAAgAXQtAEAANOVlprzYSaHw6GJEyeqR48e6tWrl5588kkVFBSUue57772n1q1bKyoqyu1jx44dvz3%2BUs2fP1/9%2BvVTly5dNHr0aO3fv/%2BS6qFpAwAAprNi0zZ9%2BnSdOnVK69at06pVq5SRkaHExMQLrt%2BlSxelpqa6fXTo0EGS9NZbb%2Bn999/X4sWL9fHHH6tJkyYaO3asnE5nheuhaQMAAPidX375RSkpKUpISFB4eLjq1q2rMWPGaNWqVSoqKrrk/SUnJ2vUqFFq3ry5goKClJCQoIyMDH333XcV3gcXIgAAANOZlYplZWUpOzvbbSwiIkJ2u93Qfnft2iVfX1%2B1atXKNdauXTudPHlSe/fudRs/6/Dhw7r//vu1c%2BdOhYSEaPz48YqLi1NBQYF%2B/PFHtW3b1rVuUFCQrr76aqWmpuqaa66pUE00bQAAwLKSk5OVlJTkNjZu3DjFx8cb2q/D4VBQUJB8fHxcY6GhoZKknJyc89YPDw9XkyZN9Oc//1ktWrTQxo0bNWXKFNntdjVr1kxOp9O1/bn7K2tfF0LTBgAATGdW0jZ48GDFxMS4jUVERFRo2zVr1mjKlCllLktISLik88369OmjPn36uL6%2B7bbbtHHjRr333nuaPHmyJF3S/spC0wYAAExnVtNmt9sv%2B1BoXFyc4uLiylz2%2BeefKy8vTyUlJfL19ZV0Jn2TpNq1a1do/w0aNNDOnTtVq1Yt2Ww21/ZnORyOCu9L4kIEAACA87Rp00ZOp1Pp6emusdTUVIWEhKhp06bnrb9y5UqtX7/ebSwjI0ONGjVSQECAWrZsqbS0NNey3Nxc7du3z3V1aUXQtAEAANNZ7ZYf4eHhio2N1YIFC3Ts2DEdOXJEixYt0sCBA%2BXnd%2BZA5ciRI12NWmFhof7yl78oNTVVRUVFWrdunTZv3qwhQ4ZIkoYOHaoVK1YoIyNDeXl5SkxMVJs2bRQVFVXhmjg8CgAAUIZZs2bp6aefVr9%2B/eTv76/bb79dCQkJruX79%2B/X8ePHJUkjRoxQfn6%2BJkyYoOzsbDVs2FCLFi1S%2B/btJUlDhgxRdna2hg8frvz8fHXt2vW8CyjK4%2BM0elYcypSdfcLTJZTLz8%2BmsLBA5eTkq7iYN4WrDMxp5WNOKxfzWfmsNKcREcEe%2B95PPGHOfufMMWe/3oikDQAAmI43jDeOc9oAAAAsgKQNAACYjqTNOJI2AAAACyBpAwAApiNpM46mDQAAmI6mzTgOjwIAAFgASRsAADAdSZtxJG0AAAAWQNIGAABMR9JmHE0bAAAwHU2bcRweBQAAsACSNgAAYDqSNuNI2gAAACyApM0kEfYQT5dQvuhoads2hfW7Xtq%2B3dPVlCvvhNPTJZTL9tufQQEBgfL392wt5QnyP%2B3pEirGx0eSFBboLzm9%2BzmQnVvo6RKuOLyWmsBDP0ckbcbRtAEAANPRtBnH4VEAAAALIGkDAACmI2kzjqQNAADAAkjaAACA6UjajKNpAwAApqNpM47DowAAABZA0gYAAExH0mYcSRsAAIAFkLQBAADTkbQZR9MGAABMR9NmHIdHAQAALICkDQAAmI6kzTiSNgAAAAsgaQMAAKYjaTOOpg0AAJiOps04Do8CAABYAEkbAAAwHUmbcTRtAADAdDRtxnF4FAAAwAJI2gAAgOlI2owjaQMAALAAkjYAAGA6kjbjaNoAAIDpaNqM4/AoAACABZC0AQAA05G0GUfSBgAAYAEkbQAAwHRWTNocDodmzJih//znP7LZbLrhhhs0ffp0Va9e/bx1p02bpjVr1riNlZSUKC4uTnPmzNHUqVO1du1a%2Bfr6upYHBARo69atFa6HpA0AAJiutNScDzNNnz5dp06d0rp167Rq1SplZGQoMTGxzHVnz56t1NRU18f27dvVrFkz3XLLLa51Hn30Ubd1LqVhk2jaAAAAzvPLL78oJSVFCQkJCg8PV926dTVmzBitWrVKRUVF5W6/fPlyXXXVVbrhhhsqrSYOjwIAANOZlYplZWUpOzvbbSwiIkJ2u93Qfnft2iVfX1%2B1atXKNdauXTudPHlSe/fudRv/vdzcXL3yyit6%2B%2B233ca//PJLbdq0ST///LOaN2%2BuGTNmqH379hWuiaYNAABYVnJyspKSktzGxo0bp/j4eEP7dTgcCgoKko%2BPj2ssNDRUkpSTk3PRbd9880116dJFLVu2dI01atRINptNEyZMUGBgoJKSkvTAAw9ow4YNCgsLq1BNNG0AAMB0ZiVtgwcPVkxMjNtYREREhbZds2aNpkyZUuayhIQEOZ3OS66npKREb731lubOnes2PnbsWLevH3vsMa1bt04pKSkaNGhQhfZN0wYAAExnVtNmt9sv%2B1BoXFyc4uLiylz2%2BeefKy8vTyUlJa4rPh0OhySpdu3aF9zn119/rcLCQnXu3Pmi39vX11f169dXVlZWhevlQgQAAIDfadOmjZxOp9LT011jqampCgkJUdOmTS%2B43aZNm9StWzf5%2Bf03F3M6nZozZ47bvgoLC7Vv3z41atSowjXRtAEAANNZ7ZYf4eHhio2N1YIFC3Ts2DEdOXJEixYt0sCBA10N2ciRI7V%2B/Xq37Xbt2qWGDRu6jfn4%2BOjAgQOaOXOmjh49qvz8fCUmJsrf31833nhjhWuiaQMAACjDrFmzFBwcrH79%2BumOO%2B5Qhw4dlJCQ4Fq%2Bf/9%2BHT9%2B3G2b7Oxs1alT57x9PfPMM2rSpInuvvtu9ejRQ7t27dLy5ctVs2bNCtfDOW0AAMB0VnxHhODgYM2bN%2B%2BCyz/66KPzxjZs2FDmurVq1dKcOXMM1XNJSdvBgwcVFRWlzMzMS/5GiYmJGj58%2BCVvZ7apU6e6dc0AAKDyWe3wqDe6pKStQYMGSk1NNasWAAAAXACHRwEAgOmqWipmhktq2g4cOKB%2B/fpp/fr1evDBB/Xoo49q48aN%2Bvrrr1W7dm3NmDFDvXr1knTmOO9zzz2nrKws3XDDDWWelHchU6dOVWBgoPz8/PTPf/5TNptNo0eP1oMPPihJatWqlZYsWaLevXtLklauXKklS5boo48%2BctX46quv6rnnntOhQ4fUv39/Pfroo5oyZYrS09MVFRWlpKQk152NnU6nZs%2BerdWrVysoKEijR4%2B%2BpEO5Zb6FRtOmsteqVeF9eETr1u6fvZzNApfNnL1xto%2BPBeo95y7fXu3cSfVyfn7e/p8u%2Bfra3D57vehoT1dQPiu9lm7f7ukKYIChpO3111/X888/r9atW2vGjBl69tlntX79euXm5iohIUGTJ0/W4MGDtWXLFk2aNElt2rSp8LqgizQAABZRSURBVL7XrVunqVOn6vPPP9fatWs1ffp0xcXFVfgGeqtXr9Y777yj9PR03Xffffr555/1wgsvqHr16rrrrru0atUqPfDAA5KkzZs367HHHtMXX3yhTz75RPHx8erYsaM6dOhQoe9V5ltoTJig%2BAkTKvx4Pep3743mrSp%2BfY3n1ajh6QoqopqnC7g0/v6erqBcYdWsM6chIZZ4kkrbtnm6goqzwmupB//4IWkzzlDT1rdvX1djExsbq9WrV6u0tFSfffaZatasqf/5n/%2BRzWbTDTfcoM6dOys/P7/C%2B27YsKHuuusuSVL//v31//7f/9NPP/1U4abtnnvuUXBwsLp06aLg4GD17NnTdQO7Dh066KeffnKta7fbNXToUEnSzTffrDZt2mjz5s0VbtrKfAuNAQOk5csrtL3HtG595kVm2DDpnBv%2BeauTn3n/i7ePz5mG7dQp6TLe/eQPVdOv0NMlVIyPz5mGrajI6yc1J7/I0yWUy9fXppCQGsrNPaWSEu//LRrW73pPl1A%2Bi72WegpNm3GGmrZzbx5XvXp1lZSUqKioSEeOHFH9%2BvVlO%2Bf4UJMmTZSWlnZZ%2B67xW2xRUFBQ4e3r16/v%2BndAQIDq1q3r9nVh4X9/YbVo0cJt28aNG%2Bvo0aMV/l5lvoXGZVxh6zHp6ZaIzK3wA3/2Ke90WqBeL2%2BAzuN0en3NxcXe/p/%2BXyUlpdao9/%2B3d%2B9BUZ33H8c/i/wQIdEaBVQkhNaMWk25OYtKqkKLWEcDZlSqrbY1yUyqMRNoTQrW2CQGexPjYGwSTdLQ/CEmxaA0oxG8ROPEy4TIJWLqLaRiuVhQjMLCLr8/MiEhoCKwHM7u%2BzWzM8s5J7ufOYH1u9/nPM8xwWdTK5N8lsK8ulW0edzgoh2bzSa73d5mm%2BM2/wW70Wt3pKPXtnyrBXyz1/v2vpaWFvXv37/T7w8AAG6uz3%2BRNQGnXInq7%2B%2BvyspKtXzjW/GZM2d67PW9vLzadN3Ky8u79XrfXneuvLy8yzefBQAAcAanFG2TJ0/W1atXtXXrVtlsNuXn5%2BvEiRM99vr33HOP8vPz1dzcrOLiYu3fv79br/ef//xH27dvV1NTkwoKCnTq1CnFxcX1TFgAAMDiuj3AKUXbsGHDtG7dOr322muyWq3asWOHFi5c2GOvn5aWpsLCQk2YMEEbNmxonQXaVfHx8Tpx4oQmTpyo5557Ts8%2B%2B6xCQkJ6KC0AAKBo6z5LS0sfv7LXrEywppTCw7%2BcTh8RYYqLZ6/W9/1fVQ8PycdHunat73%2BY3PF/jUZH6ByLRfLykmy2Pj8RofpK35%2BR6%2BnpocGDfVVb%2B4UpJiL4%2BQ80OsKtmeyz1Ki/o3HjnPO6tzHH0fS4IwIAAHC6vv5F1gx6vWh77rnntG3bthvu//Wvf62lS5f2YiIAAIC%2Br9eLtlWrVmnVqlW9/bYAAMBAdNq6j%2BFRAADgdBRt3WeSOwYDAAC4NzptAADA6ei0dR%2BdNgAAABOg0wYAAJyOTlv3UbQBAACno2jrPoZHAQAATIBOGwAAcDo6bd1Hpw0AAMAE6LQBAACno9PWfRRtAADA6Sjauo/hUQAAABOg0wYAAJyOTlv3UbQBAACno2jrPoZHAQAATIBOGwAAcDo6bd1Hpw0AAMAE6LQBAACno9PWfRRtAADA6Sjauo/hUQAAABOg0wYAAJyOTlv30WkDAAAwATptAADA6ei0dR9FGwAAcDqKtu5jeBQAAMAE6LQBAACno9PWfXTaAAAATICiDQAAOJ3D4ZyHsxUXFysuLk7z58%2B/5bFZWVmKj49XRESEFixYoJKSktZ9jY2NevrppzVlyhRFRUXp8ccfV21t7W1loWgDAABOZ8aibceOHVq%2BfLmCg4NveezevXuVmZmpP//5zzp8%2BLBiYmL06KOP6tq1a5Kk9evXq7S0VNnZ2dq9e7daWlqUmpp6W3ko2gAAADrQ2Nio7OxshYaG3vLY7OxsPfjggwoNDZW3t7cefvhhSdK%2BffvU3Nyst99%2BW0uXLtXw4cP1ne98R0888YT279%2BvysrKTudhIgIAAHA6Z3XFqqqqVF1d3Wabn5%2Bf/P39u/3a8%2BbN6/SxpaWlmjlzZuvPHh4eGjt2rIqLizV27FjV19dr3Lhxrfu/973vydvbW6WlpQoICOjUe1C0OUtLi9EJbqmqqkrZmZlK2rWrR365ne0OowN0QlVVlV59NVtJSUkmOKf9jQ7QKVVVVcrONsc59fPr%2B%2Be0qqpKmZmvmeJ8SuKz1IU4639lZma2Nm7c2GbbY489puXLlzvnDW%2Bgrq5OgwYNarNt0KBBqq2tVV1dnSRp4MCBbfYPHDjwtq5rY3jUjVVXV2vjxo3tvqGg6zinPY9z2rM4nz2Pc2qspKQk5eTktHkkJSV16r/Nzc3V6NGjO3zk5OTcdpaWW1Smt9p/K3TaAACAafn7%2B3e5w5mQkKCEhIQeyTF48ODWjtpX6urqdO%2B99%2Bquu%2B5q/dnX17d1/%2BXLlzVkyJBOvwedNgAAgG4aP368SktLW3%2B22%2B365JNPFBoaqqCgIA0aNKjN/k8//VQ2m03jx4/v9HtQtAEAAHTBjBkzdPz4cUnSggUL9M477%2Bjjjz/W9evX9be//U1eXl6aNm2a%2BvXrp/nz5%2Bull17SxYsXVVtbq4yMDMXFxWno0KGdfj%2BGR92Yn5%2BfHnvsMfn5%2BRkdxWVwTnse57RncT57HufUdcXHx6uiokJ2u10Oh0P33XefJGnXrl0KDAzUuXPnWtdhmzJlilJSUvTEE0/o0qVLuu%2B%2B%2B/TKK6/I29tbkvT444/riy%2B%2BUEJCgpqbmxUTE6M//OEPt5XH0tLdq%2BIAAADgdAyPAgAAmABFGwAAgAlQtAEAAJgARRsAAIAJULQBAACYAEUbAACACVC0AQAAmABFGwAAgAlQtAEAAJgARRvQDQsWLNDWrVtVV1dndBQAgIvjNlZoZbfb1a9fP6NjmMqLL76oPXv26MyZM4qOjtYDDzygH/3oR%2Brfv7/R0UzN4XDof//7n2w2W7t9I0aMMCCR%2BX3z79vhcKisrEzDhw/X4MGDDU4GoLMo2tzMsmXL9Mwzz2jo0KFtthcVFen3v/%2B9duzYYVAyc/v888%2B1Z88e7dmzR6dPn9aPf/xjzZ49W5MnTzY6mum8%2B%2B67euaZZ3TlypU221taWmSxWHTy5EmDkpnXoUOH9Lvf/U6HDh1Sc3Ozfv7zn%2BvUqVOSpIyMDMXExBic0LxKSkp04MABVVZWqn///goICFBcXJyCg4ONjgYXRNHmZlJSUnTo0CGtWLFC8%2BbNU0NDg9avX69t27ZpyZIlWr58udERTc1ms%2Bntt99WRkaGrl69qsDAQD3yyCP66U9/anQ004iOjtbcuXM1Y8aMDjuW3/3udw1IZW6JiYlavHixHnzwQeXm5iozM1M7d%2B5USUmJ1q5dq5ycHKMjmtKbb76pNWvWaOzYsQoKCpIklZeX69NPP9Wzzz6ruXPnGpwQroaizQ0dP35c6enp8vHx0X//%2B1%2BNGjVKK1eubP3Qwe378MMPtXPnTr333nvy9fXVrFmzlJiYqJqaGq1du1ZWq1UrV640OqYpREZG6siRI/L09DQ6issIDw/XRx99JIvFouTkZN19991KTk6WJEVEROijjz4yOKE5xcbGKjU1VXFxcW22FxQU6Pnnn9fevXsNSgZXxUQENxQWFqb4%2BHgVFxertrZWs2fPpmDroj/96U%2BaOnWqli5dqqamJm3YsEH79u3Tb3/7W40aNUoTJ07U5s2b9c477xgd1TRmzZqlo0ePGh3Dpfj6%2BurKlStqaGjQBx980DocWltbS3HcDfX19Zo2bVq77VOnTm03vA/0BP5a3cyBAweUnp6uYcOGaefOnbp48aJWr16tbdu2afXq1Qw93aaTJ08qOTlZ06dPl4%2BPT4fH%2BPv765FHHunlZOaSkZHR%2BtzHx0epqamKiIjQyJEjZbFY2hybkpLS2/FMb/bs2frFL36hfv36KSQkRGFhYWpoaNDq1asVHR1tdDzTSkhI0Pbt2zV//vw22/Py8vTAAw8YlAqujOFRNxMVFaUVK1a0udbCZrPpxRdfVFZWlgoLCw1MB3e1aNGiTh1nsViUlZXl5DSuYdasWcrLy5MkxcfHa/ny5bpy5YpmzZqlgQMHymazac2aNVqxYoXuvPNOg9Oax29%2B85vW5w6HQwcPHpS/v79GjRoli8Wic%2BfO6cKFC5o%2BfbrWrl1rYFK4Ioo2N3Pp0iUNGTKkw33//ve/de%2B99/ZyIqCtr2aJfpvdbld1dbWGDRtmQCrzmTp1qsLDw3X33Xfrtdde05IlS254LN3LzktNTe30sRRt6GkUbW7o448/Vm5uriorK7Vp0yY5HA699957mjFjhtHRAIWGhurEiRPtttfX1ys2NlbHjh0zIJX5HDlyRG%2B88YauXr2qY8eOacKECR0eR/cSMA%2BuaXMz27Zt0x//%2BEf95Cc/0cGDByVJ1dXVSk9PV1VVlRYvXmxwQrir3bt3a/fu3WpqamozBPWViooKFn%2B%2BDVFRUYqKipL05fDzP/7xD4MTuZ76%2Bnr985//1Pnz59XY2Nhmn8ViUXp6ukHJ4KqYPepmNm/erM2bN%2Bv5559vHYIKCAjQyy%2B/rDfffNPgdHBn3//%2B9zVu3DhJkpeXV7vH6NGjtXHjRoNTmhMFm3MsW7ZMr776qqqrq9XY2Njm0dDQYHQ8uCA6bW6mpqZGERERktTmuqFRo0apqqrKqFiAgoKC9NBDD8lisdz0%2BiugryguLlZ%2Bfv4NrxMGehpFm5sJDg7Whx9%2BqEmTJrXZnpeXxz0d0Sdcu3bthh01Dw8PBQQEaNKkSfy%2BwnCjR49uNywKOBMTEdzMu%2B%2B%2Bq5UrVyo2Nla7du3SggULdOrUKRUWFmrdunWKj483OiLc3JIlS1RaWqqGhgbdc8898vDw0Llz5zRgwAAFBQWppqZGNTU12rBhA/fMhKE%2B//xzpaamKjo6WgEBAfLwaHvFUWJiokHJ4Koo2txQcXGxtm/frvLycnl7eysoKEjz5s1jYV30CW%2B88YbOnDmjJ598UnfccYck6erVq/rLX/6isLAwzZkzRzk5Ofr73/%2BuHTt2GJwW7iwtLU05OTm68847O7xP7qFDhwxIBVdG0QagT4mOjlZBQYG8vb3bbG9sbNSMGTO0b98%2BORwORUZGshg0DBUWFqYtW7bccDkVoKdxTZsbWLRoUYeLlXaE9ZpgtKamJn3yySetE2a%2BcurUKdXV1UmSSkpKdNdddxkRD2g1YsQIjRkzxugYcCMUbW4gLCys9fn169eVm5uryMhIhYSEyOFw6PTp0yoqKtLChQsNTAl86Ze//KV%2B9atf6Yc//KECAwPl6empiooKHThwQHPnzpXNZtPixYuVnJxsdFS4uaeeekpPP/205s6d2%2BE1bSEhIQYlg6tieNTNpKSkKDExUVOmTGmzPT8/X3l5eXrhhRcMSgZ8bc%2BePXr//fdVXV0th8OhIUOGKCoqSgkJCbJYLDp%2B/DhDUjBcR102i8XSeiu2kydPGpAKroyizc1ERETo6NGj8vRs22RtamqS1WrlGiEA6KQLFy7cdH9gYGAvJYG7YHjUzfj7%2Bys7O1s/%2B9nP2mzPycmRn5%2BfQamAr1VUVGjLli06c%2BZMh2tgbd261YBUQHsUZehtFG1u5sknn1RKSoo2bdqk4cOHy263q7KyUvX19crIyDA6HqCUlBRdv35d999/vwYMGGB0HOCGYmNjbzrJq6CgoBfTwB0wPOqG6uvrdfDgQVVWVspms8nf31%2BTJ09WQECA0dEAhYeH68CBAxo4cKDRUYCb%2BnbX1263q7y8XPv379fDDz%2BsefPmGZQMroqizY3ZbLZ227y8vAxIAnxtzpw52rx5s4YOHWp0FKBLiouLlZmZqVdeecXoKHAxFG1u5vDhw0pPT9f58%2Bdlt9vb7We2E4x2%2BPBhZWVlaeHChQoMDGQZBZiO3W7XhAkTmNiFHkfR5mbi4uJktVo1ffr0Dq8XslqtBqQCvsYyCjCLjm5T1dDQoPz8fBUWFmr37t0GpIIro2hzM%2BHh4Tp27Fi7JT%2BAvoJlFGAWHX3B8PLyUnBwsNLS0jRp0iQDUsGV8S%2B3m7FarSorK9P48eONjgJ06Kui7NKlS7p48SK/q%2BizysrKjI4AN0Onzc289dZbev311xUTE6ORI0e2u14oKSnJoGTAlyorK5WWlqYPPvhAnp6eKikpUVVVlR566CFt2rRJQUFBRkcEWtXU1Oj8%2BfNqaGhot%2B/%2B%2B%2B83IBFcGUWbm4mNjb3hPovFwrpCMNyyZcvk4%2BOjlJQUxcfHq6ioSM3NzVq3bp3Onj2rl19%2B2eiIgCRpy5YtWr9%2BfYeTurj%2BEs5A0eYGPvvsMwUHB0uSzp49e9PFIJmZB6NFRkbq/fffl6%2Bvr0JDQ3XixAlJX17gPXXqVB05csTghMCXJk2apBUrVmjmzJny9vY2Og7cANe0uYHZs2erqKhIkjRz5swOizZm5qGvGDBggDr6Lnn58uUOOxqAURwOhxISEtSvXz%2Bjo8BNULS5gV27drU%2BZ/gTfd3EiROVlpam5ORkSdKVK1dUVlamv/71r5o2bZqx4YBvmDNnjvLy8pSQkGB0FLgJhkcB9CmXL1/WU089pf3797du8/Dw0MyZM7Vq1SoNGjTIuHDAN6xZs0b/%2Bte/NHLkyA4ndq1bt86gZHBVFG0ADHfu3Ll22y5duiQvL6/WW6v1799fEtddou9ITU296f61a9f2UhK4C4o2AIYbM2ZM610PJHHdJVzKSy%2B9pEcffdToGHABFG0ADHeruyB8E3dEgNl8cxY00B1MRABgOAoxuDJ6I%2BgpHrc%2BBAAAdNXN1sYEbgdFGwAAgAlQtAEAAJgARRsAAIAJULQBAOBETERAT6FoAwDAiebPn290BLgI1mkDAKCTFi1a1OnZoFlZWU5OA3fDOm0AAHRSWFhY6/Pr168rNzdXkZGRCgkJkcPh0OnTp1VUVKSFCxcamBKuik4bAABdkJKSosTERE2ZMqXN9vz8fOXl5emFF14wKBlcFUUbAABdEBERoaNHj8rTs%2B2gVVNTk6xWqwoLCw1KBlfFRAQAALrA399f2dnZ7bbn5OTIz8/PgERwdXTaAADogr179yolJUW%2Bvr4aPny47Ha7KisrVV9fr4yMDMXFxRkdES6Gog0AgC6qr6/XwYMHVVlZKZvNJn9/f02ePFkBAQFGR4MLomgDAKCbbDZbu21eXl4GJIErY8kPAAC64PDhw0pPT9f58%2Bdlt9vb7T958qQBqeDK6LQBANAFcXFxslqtmj59ugYMGNBuv9VqNSAVXBlFGwAAXRAeHq5jx461W/IDcBaW/AAAoAusVqvKysqMjgE3QqcNAIAueOutt/T6668rJiZGI0eOlIdH2z5IUlKSQcngqijaAADogtjY2Bvus1gsKigo6MU0cAcUbQAAdNJnn32m4OBgSdLZs2dlsVhueGxISEhvxYKboGgDAKCTfvCDH6ioqEiSNGbMmA6LtpaWFlksFpb8QI%2BjaAMAoJMqKio0YsQISdKFCxduemxgYGBvRIIboWgDAAAwAZb8AAAAMAGKNgAAABOgaAMAADABijYAAAAToGgDAAAwAYo2AAAAE6BoAwAAMIH/BzEhACRDb0uhAAAAAElFTkSuQmCC\" class=\"center-img\">\n",
       "</div>\n",
       "    <div class=\"row headerrow highlight\">\n",
       "        <h1>Sample</h1>\n",
       "    </div>\n",
       "    <div class=\"row variablerow\">\n",
       "    <div class=\"col-md-12\" style=\"overflow:scroll; width: 100%%; overflow-y: hidden;\">\n",
       "        <table border=\"1\" class=\"dataframe sample\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>y</th>\n",
       "      <th>x</th>\n",
       "      <th>length</th>\n",
       "      <th>fs</th>\n",
       "      <th>ind_numb</th>\n",
       "      <th>text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4</td>\n",
       "      <td>[22, 101, 83, 9, 27, 83, 63, 43, 38, 61, 45, 3...</td>\n",
       "      <td>30372</td>\n",
       "      <td>16000</td>\n",
       "      <td>3</td>\n",
       "      <td>a01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>6</td>\n",
       "      <td>[9, 3, 13, 13, 11, 15, 17, 15, 17, 24, 22, 22,...</td>\n",
       "      <td>25780</td>\n",
       "      <td>16000</td>\n",
       "      <td>3</td>\n",
       "      <td>a01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>[0, 1, -5, -1, -1, -1, -3, -1, -5, 0, -1, -9, ...</td>\n",
       "      <td>30045</td>\n",
       "      <td>16000</td>\n",
       "      <td>3</td>\n",
       "      <td>a01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>[23, 82, 88, -12, -114, -206, -112, -10, 59, 8...</td>\n",
       "      <td>32100</td>\n",
       "      <td>16000</td>\n",
       "      <td>3</td>\n",
       "      <td>a02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "      <td>[0, 0, 7, 0, 7, 14, 22, 22, 29, 37, 37, 14, -7...</td>\n",
       "      <td>27771</td>\n",
       "      <td>16000</td>\n",
       "      <td>3</td>\n",
       "      <td>a02</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "    </div>\n",
       "</div>\n",
       "</div>"
      ],
      "text/plain": [
       "<pandas_profiling.ProfileReport at 0x7f1565f89e80>"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "profile_file_pr = pandas_profiling.ProfileReport(data_complete_pr)\n",
    "profile_file_pr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    123\n",
       "1     79\n",
       "6     74\n",
       "4     68\n",
       "3     56\n",
       "5     43\n",
       "2     38\n",
       "Name: y, dtype: int64"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_complete_pr.y.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0     4\n",
       "1     2\n",
       "6     5\n",
       "4     3\n",
       "3    13\n",
       "5    19\n",
       "2     8\n",
       "Name: y, dtype: int64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_complete.y.value_counts() - data_complete_pr.y.value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count      481.000000\n",
       "mean     42969.288981\n",
       "std      11695.355394\n",
       "min      24976.000000\n",
       "25%      33032.000000\n",
       "50%      41443.000000\n",
       "75%      51416.000000\n",
       "max      74605.000000\n",
       "Name: length, dtype: float64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data_complete_pr.length.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Feature Engineering & Modeling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4976\n"
     ]
    }
   ],
   "source": [
    "longest_length = data_complete_pr.length.max()\n",
    "shortest_length = data_complete_pr.length.min() - 20000\n",
    "print(shortest_length)\n",
    "y_pr = data_complete_pr['y'].values\n",
    "x_pr = data_complete_pr['x'].values\n",
    "length_pr = data_complete_pr['length'].values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "def cut_into_shorter_samples(x_sample,y_sample,lenght_sample, shortest_length, threshold):\n",
    "    \"\"\"Cuts the x signal into pieces of lenght shortest_length.\n",
    "    For the last piece of signal it zero pad it. The threshold indicates at which proportion of shortest_length\n",
    "    it considers it is worth to keep the last piece.\n",
    "    Returns the list of x pieces of x, along with their respective y and length\"\"\"\n",
    "    x_array = []\n",
    "    y_array= []\n",
    "    length_array = []\n",
    "    diff_test = lenght_sample-shortest_length \n",
    "    len_to_process = lenght_sample\n",
    "    if(diff_test>0):\n",
    "        if(diff_test>shortest_length//threshold):\n",
    "            numb_new_x = lenght_sample//shortest_length\n",
    "            for i in range(numb_new_x+1):\n",
    "                if(len_to_process>shortest_length):\n",
    "                    new_x = x_sample[i*shortest_length:(i+1)*shortest_length]\n",
    "                    x_array.append(new_x)\n",
    "                    length_array.append(len(new_x))\n",
    "                    y_array.append(y_sample)\n",
    "                    len_to_process = lenght_sample-(i+1)*shortest_length \n",
    "                elif(len_to_process>shortest_length//threshold):\n",
    "                    final_x = x_sample[i*shortest_length:]\n",
    "                    len_pad = ((shortest_length-len(final_x))//2)\n",
    "                    diff_len = len_to_process+2*len_pad - shortest_length\n",
    "                    if(diff_len==0): x_pad = np.pad(final_x,pad_width =(len_pad,len_pad),mode ='constant',constant_values=0 )\n",
    "                    elif(diff_len==-1): x_pad = np.pad(final_x,pad_width =(len_pad+1,len_pad),mode ='constant',constant_values=0 )\n",
    "                    else: x_pad = np.pad(final_x,pad_width =(len_pad-1,len_pad),mode ='constant',constant_values=0 )\n",
    "                    x_array.append(x_pad)\n",
    "                    length_array.append(len(x_pad))\n",
    "                    y_array.append(y_sample)\n",
    "        else:\n",
    "            new_x = x_sample[:shortest_length]\n",
    "            x_array.append(new_x)\n",
    "            length_array.append(len(new_x))\n",
    "            y_array.append(y_sample)\n",
    "    return  x_array, y_array, length_array"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "def zero_pad_to_longest(x_pr,lengtr_pr,longest_length):\n",
    "    \"\"\"Zero pads all the x signals to match the lenght given which is longer than all the lengths.\"\"\"\n",
    "    x_final = np.zeros((len(lengtr_pr),longest_length))\n",
    "    length_array_final = [] #Just to check that indeed all are the same length\n",
    "    for i  in range(len(x_pr)):\n",
    "        len_pad = ((longest_length - length_pr[i])//2)\n",
    "        diff_len = length_pr[i]+2*len_pad - longest_length\n",
    "        if(diff_len==0): x_pad = np.pad(x_pr[i],pad_width =(len_pad,len_pad),mode ='constant',constant_values=0 )\n",
    "        elif(diff_len==-1): x_pad = np.pad(x_pr[i],pad_width =(len_pad+1,len_pad),mode ='constant',constant_values=0 )\n",
    "        else: x_pad = np.pad(x_pr[i],pad_width =(len_pad-1,len_pad),mode ='constant',constant_values=0 )\n",
    "        x_final[i,:]=x_pad\n",
    "        length_array_final.append(len(x_pad)) #Just to check that indeed all are the same length\n",
    "    return x_final,length_array_final"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "final_x = []\n",
    "final_lengths = []\n",
    "final_y = []\n",
    "for i  in range(len(x_pr)):\n",
    "    x_array, y_array, length_array = cut_into_shorter_samples(x_pr[i],y_pr[i],length_pr[i], shortest_length, 5)\n",
    "    for x_g, y_g, l_g in zip(x_array,y_array,length_array):\n",
    "        final_x.append(x_g)\n",
    "        final_y.append(y_g)\n",
    "        final_lengths.append(l_g)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "x_pr = np.array(final_x)\n",
    "y_pr = np.array(final_y)\n",
    "length_pr = np.array(final_lengths)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(4303, 4976)\n"
     ]
    }
   ],
   "source": [
    "print(np.shape(x_pr))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([1086.,  742.,    0.,  383.,    0.,  450.,  588.,    0.,  457.,\n",
       "         597.]),\n",
       " array([0. , 0.6, 1.2, 1.8, 2.4, 3. , 3.6, 4.2, 4.8, 5.4, 6. ]),\n",
       " <a list of 10 Patch objects>)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x550 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(y_pr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "def split_data_numpy(x, y, ratio=0.8, seed=1):\n",
    "    \"\"\"\n",
    "    split the dataset based on the split ratio. If ratio is 0.8 \n",
    "    you will have 80% of your data set dedicated to training \n",
    "    and the rest dedicated to testing\n",
    "    \"\"\"\n",
    "    # set seed\n",
    "    np.random.seed(seed)\n",
    "    # ***************************************************\n",
    "    index_array = np.arange(len(y))\n",
    "    np.random.shuffle(index_array)\n",
    "    final_index = int(ratio*len(y))\n",
    "    index_train = index_array[:final_index]\n",
    "    index_test = index_array[final_index:]\n",
    "    # ***************************************************\n",
    "    x_train = x[index_train]\n",
    "    y_train = y[index_train]\n",
    "    x_test = x[index_test]\n",
    "    y_test = y[index_test]\n",
    "    return x_train , y_train, x_test, y_test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train , y_train, X_test, y_test = split_data_numpy(x_pr, y_pr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train_std = np.nanstd(X_train,axis=0)\n",
    "index_std_0 = np.argwhere(X_train_std==0)\n",
    "X_train_std[index_std_0] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "dtype = torch.float\n",
    "device = torch.device(\"cuda:0\" if torch.cuda.is_available() else \"cpu\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train input:  torch.Size([3442, 4976])\n",
      "Test input:  torch.Size([861, 4976])\n"
     ]
    }
   ],
   "source": [
    "test_input = torch.from_numpy((X_test - X_train.mean(axis=0))/X_train_std).float().to(device)\n",
    "train_input = torch.from_numpy((X_train - X_train.mean(axis=0))/X_train_std).float().to(device)\n",
    "test_target = torch.from_numpy(y_test).long().squeeze().to(device)\n",
    "train_target = torch.from_numpy(y_train).long().squeeze().to(device)\n",
    "print(\"Train input: \", train_input.size())\n",
    "print(\"Test input: \", test_input.size())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train input:  torch.Size([3442, 1, 4976])\n",
      "Test input:  torch.Size([861, 1, 4976])\n"
     ]
    }
   ],
   "source": [
    "train_input = train_input.float().view(train_input.size(0), 1, -1)\n",
    "test_input = test_input.float().view(test_input.size(0), 1, -1)\n",
    "\n",
    "print(\"Train input: \", train_input.size())\n",
    "print(\"Test input: \", test_input.size())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [],
   "source": [
    "def split_data(dataset,target,k_fold):\n",
    "    \"\"\"\"\n",
    "    Split a dataset in several subsets of equal size.\n",
    "    Inputs : dataset -> original dataset\n",
    "             target -> original target\n",
    "             k_fold -> number of subsets required\n",
    "    Outputs : all_datasets -> list of all subsets of the dataset\n",
    "              all_targets -> list of subsets of the targets, matching with\n",
    "                             all_datasets\n",
    "    \"\"\"\n",
    "    N = dataset.size()[0]\n",
    "    step = N//k_fold\n",
    "    indices = torch.randperm(N);\n",
    "    all_datasets = []\n",
    "    all_targets = []\n",
    "    for k in range(0,k_fold-1):\n",
    "        all_datasets.append(dataset[indices[k*step:(k+1)*step]])\n",
    "        all_targets.append(target[indices[k*step:(k+1)*step]])\n",
    "    all_datasets.append(dataset[indices[(k_fold-1)*step:N]])\n",
    "    all_targets.append(target[indices[(k_fold-1)*step:N]])\n",
    "    return all_datasets, all_targets\n",
    "\n",
    "def build_train_and_validation(all_datasets,all_targets,k_validation,k_fold):\n",
    "    \"\"\"\n",
    "    Create a train set and a validation set from a list of subsets.\n",
    "    Inputs : all_datasets -> list of datasets\n",
    "             all_targets -> list of targets, matching with all_datasets\n",
    "             k_validation -> index in the list of the subset that should be\n",
    "                             used as the validation set\n",
    "             k_fold -> number of subsets\n",
    "    Outputs : train_data, train_target -> Train set\n",
    "              validation_data, validation_target -> Validation set\n",
    "    \"\"\"\n",
    "    assert((k_validation >= 0) & (k_validation < k_fold)),'Index of the validation set is out of bounds'\n",
    "    train = list(range(0,k_validation))+list(range(k_validation + 1, k_fold))\n",
    "    train_data = torch.cat([all_datasets[i] for i in train])\n",
    "    train_target = torch.cat([all_targets[i] for i in train])\n",
    "    validation_data = all_datasets[k_validation]\n",
    "    validation_target = all_targets[k_validation]\n",
    "    return train_data, train_target, validation_data, validation_target\n",
    "\n",
    "def train_model(model, criterion, optimizer, scheduler,train_input, train_target,n_epochs=250,batch_size=50,verbose=0,save_loss=False,tr_loss=[],val_loss=[],tr_err=[],val_err=[],val_input=torch.Tensor(),val_target=torch.Tensor()):\n",
    "    \"\"\"\n",
    "    Trains the model, using criterion as the loss function, and the optimizer wrapped in scheduler\n",
    "    Inputs: model -> the model to be trained (torch.nn.Module)\n",
    "            criterion -> loss function to be used for optimization\n",
    "            optimizer -> optimization algorithm\n",
    "            scheduler -> scheduler, changing the learning rate throughout the epochs\n",
    "            train_input -> input for training, size: (n_samples, n_channels, time_points)\n",
    "            train_target -> trarget values, size: (n_samples)\n",
    "            n_epochs -> number of passes through the whole dataset in training\n",
    "            batch_size -> the batch size for training\n",
    "            n_augmentation -> number of copies of the dataset with added noise that are to be appended to (train_input, train_target)\n",
    "            verbose -> manages output: 1 - only informs about end of training, while 2 about the current epoch % 20\n",
    "            save_loss -> a boolean indicating whether to evaluate the error& loss of the model on the whole training and validation sets\n",
    "            + lists to store results, if save_loss == True\n",
    "    \"\"\"\n",
    "    # Augment the data if necessary and make Variables\n",
    "    #train_input, train_target = Variable(train_input), Variable(train_target)\n",
    "    #if(save_loss):\n",
    "    #    val_input,val_target = Variable(val_input),Variable(val_target)\n",
    "    #train_input.requires_grad_()\n",
    "    #if(save_loss):\n",
    "    #    val_input.requires_grad_()\n",
    "    for e in range(n_epochs):\n",
    "        scheduler.step() # decrease the learning rate\n",
    "        if(verbose>=2):\n",
    "            if(e % 20 == 19):\n",
    "                print(\"Training epoch {} of {}\".format(e+1,n_epochs),end='\\r')\n",
    "\n",
    "        # permutation of the samples, for batch training\n",
    "        indices = torch.randperm(train_input.size(0))\n",
    "        train_input = train_input[indices]\n",
    "        train_target = train_target[indices]\n",
    "\n",
    "        # train on batches\n",
    "        for b in range(0,train_input.size(0),batch_size):\n",
    "            if(train_input.size(0) - b >= batch_size):\n",
    "                output = model(train_input.narrow(0,b,batch_size))\n",
    "                loss = criterion(output,train_target.narrow(0,b,batch_size))\n",
    "                #print(\"Loss: \",loss)\n",
    "                model.zero_grad()\n",
    "                loss.sum().backward()\n",
    "                optimizer.step()\n",
    "        # save loss data, if save_loss is true\n",
    "        if(save_loss):\n",
    "            tr_loss.append(criterion(model(train_input),train_target).item())\n",
    "            val_loss.append(criterion(model(val_input),val_target).item())\n",
    "            tr_err.append(evaluate_error(model,train_input.data,train_target.detach()))\n",
    "            val_err.append(evaluate_error(model,val_input.data,val_target.detach()))\n",
    "    if(verbose == 1):\n",
    "        print(\"Training ended successfully after {} epochs\".format(n_epochs))\n",
    "def evaluate_error(model, data_input, data_target):\n",
    "    \"\"\"\n",
    "    Evaluates the classification error on data_input, with respect to data_target\n",
    "    Inputs: model -> the model to evaluate performance (torch.nn.Module)\n",
    "            data_input -> input data to evaluate over\n",
    "            data_target -> the correct labels for the input data\n",
    "    Outputs: the number of missclassified samples\n",
    "    \"\"\"\n",
    "    #data_input, data_target = Variable(data_input,volatile=True),Variable(data_target,volatile=True)\n",
    "    #data_input.requires_grad_()\n",
    "    with torch.no_grad():  \n",
    "        correct = 0\n",
    "        total = 0\n",
    "        output = model(data_input)\n",
    "        temp, predicted = torch.max(output.data, 1)\n",
    "        total += data_target.size(0)\n",
    "        correct += (predicted == data_target).sum().item()\n",
    "        return (100 * (1-(correct / total))) # take the mean, to get a fraction of missclassified samples\n",
    "\n",
    "def cross_validate(model,criterion,optimizer,scheduler,dataset,target,k_fold,n_epochs=250,batch_size = 50,verbose=0):\n",
    "    \"\"\"\n",
    "    Given a model and a dataset, performs k-fold cross validation and return\n",
    "    the list of the errors on each fold.\n",
    "    Inputs : model, optimizer, scheduler, n_epochs, batch_size -> look into train_model\n",
    "             dataset, target -> dataset\n",
    "             k_fold -> number of folds to use\n",
    "    Outputs : train_errors -> list of the train error on each fold\n",
    "              validation_errors -> list of the validation error on each fold\n",
    "    \"\"\"\n",
    "    all_datasets, all_targets = split_data(dataset,target,k_fold) # prepare data for crossvalidation\n",
    "    # containers for loss\n",
    "    mean_tr_loss = [0 for i in range(n_epochs)]\n",
    "    mean_val_loss = [0 for i in range(n_epochs)]\n",
    "    mean_tr_err = [0 for i in range(n_epochs)]\n",
    "    mean_val_err = [0 for i in range(n_epochs)]\n",
    "\n",
    "    for k in range(0,k_fold):\n",
    "        if(verbose >= 1):\n",
    "            print('Cross-Validation : step {} of {}'.format(k+1,k_fold))\n",
    "\n",
    "        tr_loss = []\n",
    "        val_loss = []\n",
    "        tr_err = []\n",
    "        val_err = []\n",
    "\n",
    "\n",
    "        model.reset() # reset the model parameters\n",
    "        tr_data, tr_target, val_data,val_target = build_train_and_validation(all_datasets,all_targets,k,k_fold)\n",
    "        train_model(model,criterion,optimizer,scheduler,tr_data,tr_target,n_epochs,batch_size,verbose,True,tr_loss,val_loss,tr_err,val_err,val_data,val_target)\n",
    "        mean_tr_loss = [old + new/k_fold for old,new in zip(mean_tr_loss,tr_loss)]\n",
    "        mean_val_loss = [old + new/k_fold for old,new in zip(mean_val_loss,val_loss)]\n",
    "        mean_tr_err = [old + new/k_fold for old,new in zip(mean_tr_err,tr_err)]\n",
    "        mean_val_err = [old + new/k_fold for old,new in zip(mean_val_err,val_err)]\n",
    "        temp_tr_err = evaluate_error(model,tr_data,tr_target)\n",
    "        temp_val_err = evaluate_error(model,val_data,val_target)\n",
    "\n",
    "        if(verbose >=2):\n",
    "            print('End of step {} : Train error = {} , Validation error = {}'.format(k+1,tr_err[-1],val_err[-1]))\n",
    "\n",
    "\n",
    "    return mean_tr_loss, mean_val_loss, mean_tr_err, mean_val_err"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "class ShallowLinear(nn.Module):\n",
    "    \"\"\"2-perceptron neural net: Implements a net: \n",
    "    linear(in_features, size_hidden) -> ReLU() -> linear(size_hidden, n_classes)\"\"\"\n",
    "    def __init__(self, in_features, size_hidden, n_classes=2):\n",
    "        \"\"\"Arguments:\n",
    "            in_features - the number of features of the input\n",
    "            size_hidden - number of features in the unique hidden layer\n",
    "            n_classes - the number of classes in the classification task\n",
    "        \"\"\"\n",
    "        super(ShallowLinear, self).__init__()\n",
    "        self.in_features = in_features\n",
    "        self.size_hidden = size_hidden\n",
    "        self.n_classes = n_classes\n",
    "        print(\"Output Classes: \", self.n_classes)\n",
    "        self.lin1 = nn.Linear(self.in_features, self.size_hidden)\n",
    "        self.relu = nn.ReLU()\n",
    "        self.lin2 = nn.Linear(size_hidden, self.n_classes)\n",
    "    def forward(self, x):\n",
    "        x = x.view(-1, self.in_features)\n",
    "        x = self.lin1(x)\n",
    "        x = self.relu(x)\n",
    "        x = self.lin2(x)\n",
    "        x = torch.sigmoid(x)\n",
    "        return x\n",
    "    def reset(self):\n",
    "        self.lin1.reset_parameters()\n",
    "        self.lin2.reset_parameters()\n",
    "    def name(self):\n",
    "        return '2-perceptron neural net'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "class EEGNet_2018(nn.Module):\n",
    "    \"\"\"EEGNet : Compact CNN for EEG data. Sound data is similar as they have information local in time and several possible channels.\n",
    "    Input is: torch.Size([#Samples, #Channels, #TimePoints]), ex: torch.Size([316, 28, 500]) or torch.Size([316, 28, 50])\n",
    "    Described in :\n",
    "    V. J. Lawhern, A. J. Solon, N. R. Waytowich, S. M. Gordon, C. P. Hung, and\n",
    "    B. J. Lance. EEGNet: A Compact Convolutional Network for EEG-based\n",
    "    Brain-Computer Interfaces. CoRR, abs/1611.08024, 2016\n",
    "    https://arxiv.org/abs/1611.08024\"\"\"\n",
    "    def __init__(self, n_channels, n_time_points, n_filters=8, n_classes=2):\n",
    "        super(EEGNet_2018, self).__init__()\n",
    "        #For the last layer, the classification layer lenght\n",
    "        self.kernelL1Conv_1 = 400\n",
    "        self.kernelL2Conv_1 = 20\n",
    "        self.length = (((n_time_points//4)+1 - self.kernelL2Conv_1))\n",
    "        self.n_channels = n_channels\n",
    "        self.n_time_points = n_time_points\n",
    "        #EEGNet - Layer - 1\n",
    "        filters_layer_1 = n_filters\n",
    "        self.l1Conv2d = nn.Conv2d(1,filters_layer_1,(1,self.kernelL1Conv_1),padding = 0)\n",
    "        self.batch1 = nn.BatchNorm2d(filters_layer_1,False)\n",
    "        self.padd1 = nn.ZeroPad2d((self.kernelL1Conv_1//2,(self.kernelL1Conv_1//2)-1))\n",
    "        self.l1DeptConv2d = nn.Conv2d(filters_layer_1,filters_layer_1,(self.n_channels,1),padding = 0, groups=filters_layer_1)#Depthwise\n",
    "        self.batch1_2 = nn.BatchNorm2d(filters_layer_1,False)\n",
    "        #EEGNet - Layer - 2\n",
    "        filters_layer_2 = n_filters\n",
    "        self.conv2 = nn.Conv2d(filters_layer_1,filters_layer_2,(1,self.kernelL2Conv_1),groups = filters_layer_2)#Depthwise\n",
    "        self.conv2_2 = nn.Conv2d(filters_layer_2,filters_layer_2,kernel_size=1)#Pointwise\n",
    "        self.batch2 = nn.BatchNorm2d(filters_layer_2,False)\n",
    "        self.padd2 = nn.ZeroPad2d((self.kernelL2Conv_1//2,(self.kernelL2Conv_1//2)-1))\n",
    "        self.pool2 = nn.AvgPool2d(1,4)#(self.kernelL2Conv_1//2))\n",
    "        #EEGNet - Layer - 3\n",
    "        self.conv3 = nn.Conv2d(filters_layer_2,1,(1,self.kernelL2Conv_1))\n",
    "        #EEGNet - Clasification Layer\n",
    "        self.lin = nn.Linear(self.length, n_classes)\n",
    "        self.softMax = nn.Softmax(dim=1)\n",
    "    def forward(self,x):\n",
    "        verbose_inside = 0\n",
    "        #Reshape for the Convolutions\n",
    "        x = x.view(-1,1,self.n_channels,self.n_time_points);\n",
    "        #Layer - 1\n",
    "        if(verbose_inside): print(\"View: \",x.size())\n",
    "        x = self.l1Conv2d(x)\n",
    "        if(verbose_inside): print(\"1_l12d: \",x.size())\n",
    "        x = self.batch1(x)\n",
    "        x = nn.functional.elu(x)\n",
    "        if(verbose_inside): print(\"1_l12d_after: \",x.size())\n",
    "        x = self.padd1(x)\n",
    "        if(verbose_inside): print(\"1_padd: \",x.size())\n",
    "        x = self.l1DeptConv2d(x)\n",
    "        if(verbose_inside): print(\"1_dept: \",x.size())\n",
    "\n",
    "        x = self.batch1_2(x)\n",
    "        x = nn.functional.elu(x)\n",
    "        x = nn.functional.dropout2d(x, 0.25)\n",
    "        #Layer - 2\n",
    "        x = self.conv2(x)\n",
    "        if(verbose_inside): print(\"2_conv1: \",x.size())\n",
    "        x = self.conv2_2(x)\n",
    "        if(verbose_inside): print(\"2_conv2: \",x.size())\n",
    "        x = self.batch2(x)\n",
    "        x = nn.functional.elu(x)\n",
    "        x = self.padd2(x)\n",
    "        if(verbose_inside): print(\"2_padd2: \",x.size())\n",
    "        x = self.pool2(x)\n",
    "        if(verbose_inside): print(\"2_pool2: \",x.size())\n",
    "        x = nn.functional.dropout2d(x, 0.25)\n",
    "        x = self.conv3(x)\n",
    "        if(verbose_inside): print(\"3_conv: \",x.size())\n",
    "        #Clasification Layer\n",
    "        x = x.view(-1,self.length)\n",
    "        if(verbose_inside): print(\"cl_view: \",x.size())\n",
    "        x = self.lin.forward(x)\n",
    "        x = self.softMax.forward(x)\n",
    "        if(verbose_inside): print(\"softmax: \",x.size())\n",
    "        x = x.squeeze()\n",
    "        if(verbose_inside): print(\"final: \",x.size())\n",
    "        return x\n",
    "    def reset(self):\n",
    "        self.l1Conv2d.reset_parameters()\n",
    "        self.l1DeptConv2d.reset_parameters()\n",
    "        self.conv2.reset_parameters()\n",
    "        self.conv2_2.reset_parameters()\n",
    "        self.conv3.reset_parameters()\n",
    "        self.lin.reset_parameters()\n",
    "    def name(self):\n",
    "        return \"EEG Net (2018)\""
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Output Classes:  7\n"
     ]
    }
   ],
   "source": [
    "model =  ShallowLinear(in_features=shortest_length,size_hidden=10,n_classes=len(emotion_numbers))\n",
    "model_2 =  EEGNet_2018(n_channels=1,n_time_points=shortest_length,n_filters=8, n_classes=len(emotion_numbers))\n",
    "optimizer = torch.optim.Adam\n",
    "scheduler_gamma = 0.99\n",
    "criterion = nn.CrossEntropyLoss()\n",
    "lr = 1e-3\n",
    "epochs =75\n",
    "cross_val = True\n",
    "model_list = [model,model_2]\n",
    "optimizer_list = [optimizer]*len(model_list)\n",
    "scheduler_gamma_list = [scheduler_gamma]*len(model_list)\n",
    "criterion_list = [criterion]*len(model_list)\n",
    "lr_list = [lr]*len(model_list)\n",
    "epochs_list = [epochs]*len(model_list)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Results & Visualizations"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "cross_val = True\n",
    "save_loss = True"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model 2 of 2 : EEG Net (2018)\n",
      "Cross validating... on 4 folds\n",
      "Cross-Validation : step 1 of 4\n"
     ]
    }
   ],
   "source": [
    "dump_final = []\n",
    "if(cross_val):\n",
    "    dump = []\n",
    "\n",
    "for i in range(1,len(model_list)):\n",
    "    model = model_list[i]\n",
    "    optimizer = optimizer_list[i](model.parameters(),lr=lr_list[i])\n",
    "    scheduler = torch.optim.lr_scheduler.ExponentialLR(optimizer,scheduler_gamma_list[i])\n",
    "    criterion = criterion_list[i]\n",
    "    print(\"Model {} of {} : {}\".format(i+1,len(model_list),model.name()))\n",
    "    if(cross_val):\n",
    "        k_fold = 4\n",
    "        print(\"Cross validating... on {} folds\".format(k_fold))\n",
    "        tr_loss, val_loss, tr_err, val_err = cross_validate(model, criterion,optimizer,scheduler,train_input,train_target,k_fold,batch_size=5,n_epochs=epochs_list[i],verbose=2)\n",
    "        print(\"Mean train error : {}, mean validation error : {}\".format(tr_err[-1],val_err[-1]))\n",
    "        dump.append((model.name(), \" train \", tr_loss, tr_err))\n",
    "        dump.append((model.name(), \" validation \", val_loss, val_err))\n",
    "\n",
    "    model.reset()\n",
    "    print(\"Training...\")\n",
    "    train_model(model,criterion,optimizer,scheduler,train_input,train_target,n_epochs=epochs_list[i],batch_size=5,verbose=2)\n",
    "    final_tr_error = evaluate_error(model,train_input,train_target)\n",
    "    final_te_error = evaluate_error(model,test_input,test_target)\n",
    "    print(\"Train error = {} ; Test error = {} \".format(final_tr_error,final_te_error))\n",
    "\n",
    "    dump_final.append((model.name(), \" train \", final_tr_error))\n",
    "    dump_final.append((model.name(), \" test \" , final_te_error))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "file = open('final_'+str(3)+'t.txt','w+')\n",
    "json.dump(dump_final,file)\n",
    "file.close()\n",
    "\n",
    "if(cross_val):\n",
    "    file = open('cv_'+str(3)+'t.txt','w+')\n",
    "    json.dump(dump,file)\n",
    "    file.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 1.0 models to plot in this file.\n",
      "Print the models all together\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x13888374f28>"
      ]
     },
     "execution_count": 117,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 800x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "f_name = 'cv_3t.txt'\n",
    "file= open(f_name,'r')\n",
    "\n",
    "lines = json.load(file)\n",
    "n_models = len(lines)/2\n",
    "print(\"There are \"+str(n_models)+\" models to plot in this file.\") #should be 4\n",
    "print(\"Print the models all together\")\n",
    "colors = ['r','r','b','b','g','g']\n",
    "fig1 = plt.figure(figsize=(8,8))\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Loss (Cross entropy)')\n",
    "for i in range(0,len(lines),2):\n",
    "    model = lines[i][0]\n",
    "    tr_loss = lines[i][2][2:]\n",
    "    val_loss = lines[i+1][2][2:]\n",
    "    epochs = range(2,len(tr_loss)+2)\n",
    "    plt.plot(epochs,tr_loss,label=model+', train',color=colors[i],linestyle='--')\n",
    "    plt.plot(epochs,val_loss,label=model+', validation',color=colors[i+1])\n",
    "plt.legend()\n",
    "#lt.savefig(path+\"loss_all_\"+str(n_time_points)+\".png\")\n",
    "fig2 = plt.figure(figsize=(8,8))\n",
    "plt.xlabel('Epoch')\n",
    "plt.ylabel('Classification error rate')\n",
    "for i in range(0,len(lines),2):\n",
    "    model = lines[i][0]\n",
    "    tr_err = lines[i][3][2:]\n",
    "    val_err = lines[i+1][3][2:]\n",
    "    epochs = range(2,len(tr_err)+2)\n",
    "    plt.plot(epochs,tr_err,label=model+', train',color=colors[i],linestyle='--')\n",
    "    plt.plot(epochs,val_err,label=model+', validation',color=colors[i+1])\n",
    "    plt.ylim((0,1.0))\n",
    "plt.legend()\n",
    "#lt.savefig(path+\"classErr_all_\"+str(n_time_points)+\".png\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Training...\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n",
      "View:  torch.Size([10, 1, 1, 4976])\n",
      "1_l12d:  torch.Size([10, 8, 1, 4577])\n",
      "1_l12d_after:  torch.Size([10, 8, 1, 4577])\n",
      "1_padd:  torch.Size([10, 8, 1, 4976])\n",
      "1_dept:  torch.Size([10, 8, 1, 4976])\n",
      "2_conv1:  torch.Size([10, 8, 1, 4957])\n",
      "2_conv2:  torch.Size([10, 8, 1, 4957])\n",
      "2_padd2:  torch.Size([10, 8, 1, 4976])\n",
      "2_pool2:  torch.Size([10, 8, 1, 1244])\n",
      "3_conv:  torch.Size([10, 1, 1, 1225])\n",
      "cl_view:  torch.Size([10, 1225])\n",
      "softmax:  torch.Size([10, 7])\n",
      "final:  torch.Size([10, 7])\n"
     ]
    },
    {
     "ename": "KeyboardInterrupt",
     "evalue": "",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-58-2ea4dbda0166>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      6\u001b[0m \u001b[0;31m#model.reset()\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m      7\u001b[0m \u001b[0mprint\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m\"Training...\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 8\u001b[0;31m \u001b[0mtrain_model\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mmodel\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mcriterion\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0moptimizer\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mscheduler\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mtrain_input\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mtrain_target\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mn_epochs\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m2\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mbatch_size\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m10\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0mverbose\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;36m2\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m<ipython-input-37-27aae6f1f6d5>\u001b[0m in \u001b[0;36mtrain_model\u001b[0;34m(model, criterion, optimizer, scheduler, train_input, train_target, n_epochs, batch_size, verbose, save_loss, tr_loss, val_loss, tr_err, val_err, val_input, val_target)\u001b[0m\n\u001b[1;32m     82\u001b[0m                 \u001b[0mmodel\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mzero_grad\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     83\u001b[0m                 \u001b[0mloss\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msum\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mbackward\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 84\u001b[0;31m                 \u001b[0moptimizer\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mstep\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     85\u001b[0m         \u001b[0;31m# save loss data, if save_loss is true\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     86\u001b[0m         \u001b[0;32mif\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0msave_loss\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/dl_epfl/lib/python3.6/site-packages/torch/optim/adam.py\u001b[0m in \u001b[0;36mstep\u001b[0;34m(self, closure)\u001b[0m\n\u001b[1;32m    104\u001b[0m                 \u001b[0mstep_size\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0mgroup\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'lr'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m*\u001b[0m \u001b[0mmath\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0msqrt\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mbias_correction2\u001b[0m\u001b[0;34m)\u001b[0m \u001b[0;34m/\u001b[0m \u001b[0mbias_correction1\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    105\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 106\u001b[0;31m                 \u001b[0mp\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mdata\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0maddcdiv_\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m-\u001b[0m\u001b[0mstep_size\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mexp_avg\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdenom\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    107\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    108\u001b[0m         \u001b[0;32mreturn\u001b[0m \u001b[0mloss\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
     ]
    }
   ],
   "source": [
    "i=1\n",
    "model = model_list[i]\n",
    "optimizer = optimizer_list[i](model.parameters(),lr=lr_list[i])\n",
    "scheduler = torch.optim.lr_scheduler.ExponentialLR(optimizer,scheduler_gamma_list[i])\n",
    "criterion = criterion_list[i]\n",
    "#model.reset()\n",
    "print(\"Training...\")\n",
    "train_model(model,criterion,optimizer,scheduler,train_input,train_target,n_epochs=2,batch_size=10,verbose=2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train error = 85.50261475886113 ; Test error = 86.99186991869918 \n"
     ]
    }
   ],
   "source": [
    "final_tr_error = evaluate_error(model,train_input,train_target)\n",
    "final_te_error = evaluate_error(model,test_input,test_target)\n",
    "print(\"Train error = {} ; Test error = {} \".format(final_tr_error,final_te_error))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1225"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "shortest_length//4 +1 - 20"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "18652"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}