beijing_hotels.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Парето-оптимальность, Selenium и отели в Пекине"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Название [моего блога](https://vk.com/mat.chast) как бы намекает, что его содержание применимо в быту. К сегодняшнему блокноту это точно относится: он про выбор отелей. \n",
    "\n",
    "Моя подруга на новогодние праздники собралась в Пекин, и для этого решила выбрать отель. В Пекине их тысяча, и она ограничила поиск только пятизвёздочными. Но даже таких отелей там больше сотни, и из них хочется выбрать наилучший по соотношению цена-качество.\n",
    "\n",
    "Сколько можно заплатить за дополнительный балл оценки (она измеряет качество)? На этот вопрос ответ у каждого свой. Но чтобы не перебирать вручную 150 отелей, можно заставить алгоритм отобрать с десяток отелей, каждый из которых был бы лучшим в своей ценовой категории. В математике такие отели называются Парето-оптимальными.\n",
    "\n",
    "А раз уж мы всё равно пользуемся алгоритмом, почему бы не распарсить booking.com тоже полностью автоматически?\n",
    "\n",
    "Именно этому - парсингу booking.com и выбору оптимального отеля - и посвящён блокнот. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# Сбор данных при помощи Selenium"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Предполагается, что:\n",
    "* у вас установлены все необходимые питоновские пакеты. Если какой-то импорт не работает, зовите pip install\n",
    "* у вас установлен Firefox (если не хотите его ставить, попробуйте другой бразуер, но к нему нужен свой драйвер)\n",
    "* у вас есть geckodriver, скачанный отсюда https://github.com/mozilla/geckodriver/releases и распакованный\n",
    "* этот geckodriver находится в той же папке, что и блокнот (либо добавлен в PATH)\n",
    "\n",
    "Если вам лень заниматься всей этой технической частью, можете [скачать данные](https://github.com/avidale/weirdMath/blob/master/hotels.pkl) и сразу перейти к их анализу. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Импортируем Селениум. Если эта ячейка выдаёт ошибку, вызовите pip install selenium в командной строке и перезапустите блокнот"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "from selenium import webdriver"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Запускаем браузер в управляемом режиме. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "driver = webdriver.Firefox() "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ниже - ссылка на запрос отеля в Пекине с ограничением на даты и звёздочки"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "url = \"https://www.booking.com/searchresults.ru.html?label=gen173nr-1FCAEoggJCAlhYSDNYBGjCAYgBAZgBIcIBCndpbmRvd3MgMTDIAQzYAQHoAQH4AQuSAgF5qAID&sid=6b8e3d83b69499330279412a904e1bd4&checkin_month=1&checkin_monthday=4&checkin_year=2018&checkout_month=1&checkout_monthday=11&checkout_year=2018&class_interval=1&dest_id=-1898541&dest_type=city&dtdisc=0&from_sf=1&group_adults=2&group_children=0&inac=0&index_postcard=0&label_click=undef&no_rooms=1&oos_flag=0&postcard=0&room1=A%2CA&sb_price_type=total&search_pageview_id=2f838df472d401ec&src=index&ss=Пекин&ss_all=0&ssb=empty&sshis=0&nflt=class%3D5%3B&lsf=class%7C5%7C124&unchecked_filter=class\""
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Сейчас будет магия: открытый из кода браузер сам перейдёт по ссылке"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "driver.get(url)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Напишем функцию, которая бы находила внутри блока с отелем его название, рейтинг и цену"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def get_data_from_hotel(div):\n",
    "    name = div.find_elements_by_css_selector('span.sr-hotel__name')[0].text\n",
    "    \n",
    "    rating_elements =  div.find_elements_by_css_selector('span.review-score-badge')\n",
    "    if rating_elements:\n",
    "        rating = rating_elements[0].text\n",
    "    else:\n",
    "        rating = 0\n",
    "    \n",
    "    price_elements = price = div.find_elements_by_css_selector('strong.price')\n",
    "    if price_elements:\n",
    "        price = price_elements[0].text\n",
    "    else:\n",
    "        price = None\n",
    "    \n",
    "    hotel = {'name':name, 'price': price, 'rating': rating}\n",
    "    return hotel"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Теперь пройдёмся по всем 15 блокам с отелями и выкачаем цены. \n",
    "\n",
    "Для двух отелей цен нет: все комнаты уже забронировали."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "hotels = []"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'name': 'New World Beijing Hotel', 'price': '46 554 руб.', 'rating': '9,1'}\n",
      "{'name': 'DoubleTree by Hilton Beijing', 'price': '35 153 руб.', 'rating': '7,5'}\n",
      "{'name': 'Sofitel Wanda Beijing', 'price': '55 237 руб.', 'rating': '8,3'}\n",
      "{'name': 'Ascott Raffles City Beijing', 'price': '69 792 руб.', 'rating': '8,8'}\n",
      "{'name': 'Artyzen Habitat Dongzhimen Beijing', 'price': '43 488 руб.', 'rating': '8,5'}\n",
      "{'name': 'Crowne Plaza Beijing Lido', 'price': '61 820 руб.', 'rating': '8,4'}\n",
      "{'name': 'Hotel Éclat Beijing', 'price': '92 874 руб.', 'rating': '9,1'}\n",
      "{'name': \"W Beijing Chang'an\", 'price': '85 549 руб.', 'rating': '8,5'}\n",
      "{'name': 'Conrad Beijing', 'price': '57 686 руб.', 'rating': '8,4'}\n",
      "{'name': 'The Ritz-Carlton Beijing', 'price': '106 825 руб.', 'rating': '8,8'}\n",
      "{'name': 'DuGe Boutique Hotel', 'price': '50 307 руб.', 'rating': '7,9'}\n",
      "{'name': 'JW Marriott Hotel Beijing', 'price': '82 878 руб.', 'rating': '8,3'}\n",
      "{'name': 'Crowne Plaza Beijing Zhongguancun', 'price': '50 884 руб.', 'rating': '8,1'}\n",
      "{'name': 'Shangri-la Hotel Beijing', 'price': '54 753 руб.', 'rating': '8,1'}\n",
      "{'name': 'Shangri-La Kerry Hotel, Beijing', 'price': '77 893 руб.', 'rating': '9,0'}\n"
     ]
    }
   ],
   "source": [
    "for div in driver.find_elements_by_css_selector('div.sr_item'):\n",
    "    hotel = get_data_from_hotel(div)\n",
    "    print(hotel)\n",
    "    hotels.append(hotel)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Теперь всё то же самое проделаем для следующих нескольких страниц с отелями. \n",
    "\n",
    "Для этого заставим браузер находить кнопку \"Следующая страница\", жамкать на неё самостоятельно, и ждать 3 секунды, пока отели подгрузятся. \n",
    "\n",
    "Заодно будем печатать, сколько отелей показано на странице. \n",
    "\n",
    "Если в какой-то момент кнопку не удастся найти, это значит, что все отели уже выкачаны. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Пекин: найдены 124 уникальных варианта жилья показаны 1 – 15\n",
      "Пекин: найдены 124 уникальных варианта жилья показаны 16 – 45\n",
      "Пекин: найдены 124 уникальных варианта жилья показаны 46 – 75\n",
      "Пекин: найдены 124 уникальных варианта жилья показаны 76 – 105\n",
      "Пекин: найдены 124 уникальных варианта жилья показаны 106 – 124\n"
     ]
    }
   ],
   "source": [
    "import time\n",
    "\n",
    "for page in range(100):\n",
    "    bottom = driver.find_elements_by_css_selector('div.results-meta')[0]\n",
    "    print(bottom.find_elements_by_css_selector('h4')[0].text)\n",
    "    buttons = driver.find_elements_by_css_selector('a.paging-next')\n",
    "    if buttons:\n",
    "        buttons[0].click()\n",
    "        time.sleep(3)\n",
    "        for div in driver.find_elements_by_css_selector('div.sr_item'):\n",
    "            hotel = get_data_from_hotel(div)\n",
    "            hotels.append(hotel)\n",
    "    else:\n",
    "        break"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Ура! Нам удалось выкачать все пятизвездочные отели в Пекине!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "124\n"
     ]
    }
   ],
   "source": [
    "print(len(hotels))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Анализ данных"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Теперь, когда у нас есть данные по отелям, осталось визуализировать их и найти \"лучшие\" в каком-то смысле. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Если вы не собирали данные Селениумом, самое время скачать их [отсюда](https://github.com/avidale/weirdMath/blob/master/hotels.pkl) и положить в ту же папку, что и блокнот. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Если вы собрали данные, пропустите эту ячейку. Иначе скачайте их и попробуйте снова.\n"
     ]
    }
   ],
   "source": [
    "import pickle\n",
    "try:\n",
    "    hotels = pickle.load(open('hotels.pkl', 'rb'))\n",
    "except FileNotFoundError:\n",
    "    print('Если вы собрали данные, пропустите эту ячейку. Иначе скачайте их и попробуйте снова.')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Классический импорт библиотек"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Кладём данные в удобную таблицу pandas"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "      <th>price</th>\n",
       "      <th>rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>New World Beijing Hotel</td>\n",
       "      <td>46 554 руб.</td>\n",
       "      <td>9,1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>DoubleTree by Hilton Beijing</td>\n",
       "      <td>35 153 руб.</td>\n",
       "      <td>7,5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Sofitel Wanda Beijing</td>\n",
       "      <td>55 237 руб.</td>\n",
       "      <td>8,3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Ascott Raffles City Beijing</td>\n",
       "      <td>69 792 руб.</td>\n",
       "      <td>8,8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Artyzen Habitat Dongzhimen Beijing</td>\n",
       "      <td>43 488 руб.</td>\n",
       "      <td>8,5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 name        price rating\n",
       "0             New World Beijing Hotel  46 554 руб.    9,1\n",
       "1        DoubleTree by Hilton Beijing  35 153 руб.    7,5\n",
       "2               Sofitel Wanda Beijing  55 237 руб.    8,3\n",
       "3         Ascott Raffles City Beijing  69 792 руб.    8,8\n",
       "4  Artyzen Habitat Dongzhimen Beijing  43 488 руб.    8,5"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.DataFrame(hotels)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Сейчас цена и рейтинг - это текстовые строки. Превратим их в числа:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "data.rating = data.rating.str.replace(',', '.').astype(float)\n",
    "data.price = data.price.str.replace('руб.', '').str.replace(' ', '').astype(float)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Теперь данные можно визуализировать"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAElCAYAAAALP/6mAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzt3XucHFWZ//HPN8kEEgQSYBAId8Eg\niNyyXBYvICqCIIiooKyCqywuqLhr/IEioKuContBVgKCV1jkahYFRHYBrwtsQoCIkBUQSCYIAyHh\nkgEmyfP745zu1NRUdVf3dPX0TD/v16tf013XU9U1fbrPU+c5MjOcc845gAmjXQDnnHOdwysF55xz\nVV4pOOecq/JKwTnnXJVXCs4556q8UnDOOVfllYJzzrkqrxScc85VeaXQpSQ9KultidczJD0i6Ruj\nWS7n3OjySsEhqRf4L+AXZva50S6Pc270eKXQ5SRNA34J3AWcnJi+t6T/kbRc0hOSLpA0OTHfJO0Q\nn79b0mJJ2+fs43ZJH4vPJ0haKGlJjTLtIukWScskPSnp84l5x0taLemF+EiWY0NJl8by9kn6iqSJ\niXV3iMtX1l1dKVfiOO6Px3y7pNcl5j0qaSCu1yfplMS8EyQ9IOn5+Gvr71LHs23efiWtI+lfJS2N\nj3+VtE7OeZkg6QxJj0l6StKPJG2YOLbFkvZPnKffxudTJP1e0qfj6wOS51/S+2P5kuei3rm6WtJf\nJK2Q9GtJu8TpH0it81LldZx/tqTL8t57N/q8UuhurwJuAiYBH7WhibBWA58BNgH2Aw4C/j69AUlv\nAeYAh5rZIwX2+RFget5MSesTf7UAWwA7AP+dWGQC8Hsze5WZvSq1+g+BVXGdPYB3AB9LrUti3d8k\n9vta4ArgVKAXuBH4WbIiBA6P630QOF/SBnH6U8BhwAbACcC/SNozvV9gw/R+gS8A+wK7A7sBewNn\n5Jye4+PjQGB7wvt3QTymh4D3AT+RtGPiuCYAPwbuMrN/S29QUg/wT8ATqVm55yq6CdgR2BS4G7g8\nLn9lap1Tct4r16G8UuhuFwIvAFsC+ydnmNl8M7vDzFaZ2aPARcBbUuvvAVwPfMjMFtbbmaR1gS8S\nPoTyHAb8xcy+ZWYvmdnzZnZnYv5k4JWMbb8aOAQ41cxeNLOngH8Bjkkstm7WutEHgBvM7BYzGwS+\nCUwB/jpj2UnAc5VtmdkNZvawBb8i/PJ6U6rMa8xsdca2PgR82cyeMrN+4EvA3+SU8UPAP5vZI2b2\nAnA6cIykSbEcdwBnEj6we+M63yR8cP9Dzjb/DrgT+L/U9FrnCjP7XnxvXgbOBnar/GpxY5tXCt3t\nQeBw4HPApZKmVGZIeq2kn8cmgueArxF+NSRdAvwJeHvB/X0auBlYVGOZrYCHa8zfCHg2Y/o2QA/w\nRGz+WU6oyDZNLLMZ0J+z3S2AxyovzGwNsBiYkVhmbjwXvwS+ZmYvAUg6RNIdsblrOXAoQ89VXpmH\n7Tc+36KBZScBr05MezuwDPgysBfhF95Mhp4HYrnXJ7z3X8zYV+65kjRR0rmSHo7n49E4K3195Hl/\nfI+ejs2Emc2ObnR4pdDdvhq/jX8XeJyh3+AvJFQaO5rZBsDnAaXWP5Xwzf5vU80lWTYCTiF8E65l\nMfCaGvNfy/BvtZX1XgY2MbNp8bGBme2SWGYP4N6c7S4lVCwASBKhgupLLHNkPBdbA5+WtF9s/7+W\n8I381WY2jdD0lDxXeWUett+47aUNLLsKeDKW+e3ALODNhCaj5YRK4kLgXzO2Nxu4yswey5hX61x9\nEDgCeBuwIbBtnJ6+PvJcFc/TFoTr7msF13Nt4JWCq/g4cKKkvePr9QlNJC9I2gn4RMY6vzGzvwCf\nBb4f26fznApcGpev5efAZpJOjUHY9SXtAxCDqEcC/5leycyeIHyD/5akDWJQ9jUx5kFs/z+eEDfI\nchXwLkkHxeP4R0Il8/uMZSvNQL2EpqF1CN+qV0k6hBDLIO53K8IvpLk5+70COENSr6RNCM0/eYHY\nK4DPSNpO0qsIH6ZXmtmq2DT3HeDv4y+Y3wMPx2a0c4DdJb0zsa31CfGPr6Z3UuBcrU84N88AU2ny\nQ93MXiE0X06I+60E5LdtZnuuNbxScADEIPGZhA/3yYQP+g8CzwPfBa6sse6PCd/UP5+3DDCR8G26\nXjmeJ3y7PRz4C6F56kBJOxMCyZ9NxRiSPkz4kP4jobnmGmDzOG8esBNwUeJumDcBF0ja2swWAccB\n3waejvs/PH5wVfwsrncfcB0hBvE88ClCpfIs4Zxdn1jnZuB2Qnwjy1di2e4DFhKCtl/JWfZ7hF8A\nvwb+DLwEfDLOOwO4w8z+K71SbPc/Cfj3RBPhBsD5ZpbVrFXzXAE/IjRd9RHO9R055c3zHklLJPUB\ne7I2sL5VYrtulMhHXnPdQNKjZrZtxvRLgK/EYLpj9M6VpDOAfjO7qIztu2ImjXYBnGuT9C2XFcsI\n7fJurVE5V2aW9wvJtZH/UnDOOVflMYUa6gROnXNu3PFKISHef32apHsl/QVYEu/qcF1M0l9L2l9S\nj6SPS5pa4r4+KGlLhZQdHy9rP87l8UphqH8n3PnyHjPbzMxeXemg5LraM4R7/Z8E3mhmK0vc18vA\n/xA6+E2us6xzLecxhUjSdoRb63aItxk651z3MbNx+yB88/9WatrPCPlx0su+H7gHuA1YAdwPvDsx\nf0PC/dn9hHupzwAmJOYfAKwhdMZ5IT5/W5y3DuEe/ccJ3zbnAFMS6x4W972c0OnoDXWO67eEe9Rf\niH9/m5i3BeE++WXAQ8DHE/POJty7fyWh/8HdwG6J+a8j3FO/POP4f0DIhfNC4rFhPO4lqfNowMfi\n6wuBaxLzv05IcKeM45oQz+tjhCRzPyIkkYPQu/YFYCB1nj8f5+8E3BKPexHw/tS2zwYG4zovxjJO\nSpyza+N7+2fgU6n1Lku8/k5cd4ec9+Z2Qmexu+J19J/ARon5+8b3eHk8pgNS634scS4WVs4tIfFd\n5ZgtHsMLwE311o3THiVej/H1x4DbE6+T23wBuKLG8VWuvacIveLzrtONgO8TemI/C8xN/K8sIfRr\neTqW7UOJ9er9v0xKlXeQcKts8lh2SLz+CvCDBt6DV4BNE9OuKvCeZ54Twv/NVzLWyTr/O8TnOwCW\nmPc24KHU/3jm9dqKx3hvPvohcGzMFEnsMXoQ2T01pxKyVN5C6Kn6SeBySTPj/G8TPgS3JySG+zCh\nR2jFBKDP1maEfDwx7+uEVAe7E97wGYSOYsT0EN8jJCbbmJCv53rlpE+OBPxd3M9JqXlXEP7htgCO\nBr4m6aDE/COAqwn/sP9ByOfTE4PqPyP0Ct404/gBvlE5vvhYMaRQ2Rk3/xF4g0Iq5zcBfwt8xOLV\nnXI8+VlAd4vHewiwNFGGr0laj/C+/Ucs+7HAdyrpnKMJwE/iNqrT47XxM8KHwwzC9XGqpIPThYvZ\nRw/JKHfah4GPEt6DVcD5cf0ZwA2ED6mNCB0Er1UYzyJtSDZZM0tnG90tvs4qT81MtDXsljivx9ZY\n7pRYjjcC/yjp9TnL/Zjwf7UL4X1JduDbjJAraUYs78WJay33/yWqfG7tHMtxedEDLPgePBzLVPnM\neG2BTRc9JyPSyPXarHFdKZhZ5dta5UPxGELt/GTOKk8C55rZK2Z2KyHlwrEKOfk/AJxuITPko8C3\nGJrNMi97pwgpJD5jZsssNE19jbXZOz8OXGRmd5rZajP7IaFded8ahzYlZ19bES7K/2chp9E9hKR1\nyXLON7NrLGQC/WdCNsx94+NVWcdfoxxpwzJuWmh/Py7u6zLgk2aWN5ZCzSygNRwGPGpm37eQ1fVu\nwjepoxPLZL4/wF8BvWb25XjcjxB6cB+Tsew51M7wWvFjM/uDmb1ISDb3/ngNHQfcaGY3mtkaM7uF\n0Hv40OTKBbPJZhrJuk2YREj5sSI9Q9LmhAr0JDN71swGLWSQTfqimb0cp99AOE/1/l8gXLNQI4tr\nDUXegx+x9n/mw4TKrajcc9IijVyvTemGzms/JFwIt8S/w3LKRy8Diy1kx6x4jFAbb0L4UElnqExm\n0MzLhNlL+LY0P1zvQPimXxn8ZRvgI5I+mVhnMvmZMiE/g+UWQOUfKVnOWYnXiytPzGyNwmArlX3l\nHX9diYybbyKc8yozu0vSI4Rvi1fV2EytLKC1Uh9sA+wTM5RWTGLoP3Ot7KpbpNadSGr8gJh/aSfC\nl4NLa5QFEuc4HkMP4RraBnifpMMT83sITZZJRbLJ5qm17lxJlc5nkwlNXM04X9I3Cb+cLzCzxRnL\nbEW4FvOywz4bK82KSnbYev8vEK7/NYQbAPLcLalyLa8L/CQ+L/Ie9AP/F3/Z/g0hl1W9FC21zsln\nFQZlehn4HaHSa1ah63UkxvUvhegy4AhJuxHazPMSkz0ObFVpaoq2JnwYPU1ot0xnqEx+UOVlwnya\n0A6+i63N3lkZbAXCB8hXE/OmmdlUM8tMRiZpM8I/xX0Zs5cCG8UP6LxybpXY1gTCWApL4yPv+IvI\nzbgp6WRCO/FSQsWRp2YW0BoWA79KncNXmVkyiV+t7Kp/Tq27vpkdmlruG8Bplj0mQtpWiedbE66d\np+O+fpza13pmdm5i+aLZZLPUW/fIyn4J+Zqa9am4jY2AN0rK+jW5mHAtTsvZxvTY7FdRyQ5b7/8F\nQgbXB21oXqq0PRPHmvxAL/IeQPiF/W1CW35euvWkWufkm3He9oQKb3aB7eUper02bdxXCrGp4n8J\n3xqvNbOBnEXvJASuPhfb2A8gJEX7SfwguAr4aszauQ1h0JLLAGKyto+SUeHEb97fJYzGtWlcfkai\nDfC7wEmS9lGwnqR3pT7Ykz4F3Goh+2V6X4sJAbRzJK0r6Q2ENvxkm+teko6KTTKnEr693FHr+HPK\nkVQr4+ZrCe23xxG+dX1O0u4528nNAlpn/z8HXivpbyrxEUl/Jel18ZweQfi1dFPGuncBz0n6fwrD\nVk6U9HpJf5VY5q2EwN/P65Sj4jhJO8f+DF8mBNpXE66XwyUdHPezrsLQmFsm1i2aTTbLSNZtxmpC\ngHRYTMRC1tqbCLGd6fE9eXNqsS9Jmhy/kR8GXF3v/0UhWeNJ5GdwrafIewAhtnY3+YkM8+SeE0Iw\neiUj+9wtcr2OjLUwat2pD8IHkgEH1lluF0IGysrdR0cm5k0nXFD9hNr6TMKbux4hsPuF1LYeZe3d\nR+sSPuAeIaSjfoChd7i8k1BxLScEaa8G1s8o3+fjcbzM2rtEXiJciJW7cLYkfEguIwTMTkqsfzZD\n7z5aQPhGlTz+X8Xj/yOhv0a9uygOiGWanZh2O+HuikmEi/i0xLxPEO6MWSdjWxPieV0cz/NlwPSM\n/S3JWHcmoV26n9CscCshUHlI3N8RiWW3ZfjdR1cQsrI+S6gk35Y4ZwbsnVi/6N1HzxGCgpsk5u8T\nz/GyWNYbgK0T6z4JvKrOsQ7bf711KXb3UeYxZeyncqfNMkJwf72cZTciNCU+Gc/rdcmyEYYifZrw\nK/1vEuvl/r8QPqyN8OFa+R8YJMQX3pR1LAy/+6jee/CxIue8yDkh/N88F4+3jzDWxqY55/+JuNwT\n8fWS+Ohn+N1HmddrKx5d0U8hfkO5DNjWhraZjymSziYEVH+Qmv5GwkVxdoH1dzCz40oqYteTdDvh\nFtZLRrssnSr+Cr3MzNLfzouseztwvKUytSpkWP2tmd3egiJ2tXHffKRwm+SngUvGcoUQPUdo4kl7\nOc5zbrzrJztT63OE/wM3QuP67iNJryPcbnYvQ/sUjElm9s850/+X0Pzk3LhmZu/LmX5+u8syXnVF\n85Fzzrlixn3zkXPOueLGXPPRJptsYttuu+1oF8M558aU+fPnP21mWbfKDjHmKoVtt92WefPmjXYx\nnHNuTJE0rGNpFm8+cs45V+WVgnPOuSqvFJxzzlV5peCcc67KKwXnnHNVXik455yrKrVSkPRpSX+Q\ndL+kUzPmHyBphaR74uPMrO0410pzF/Sx/7m3st1pN7D/ubcyd0HRISOcG/9K66egMEbpx4G9CWlt\nfyHpBjP7U2rR35jZYWWVw7mkuQv6OP26hQwMhrFy+pYPcPp1CwE4co9Cg8w5N66V+UvhdcAdZrbS\nwiApvwLeU+L+nKvrvJsXVSuEioHB1Zx3czMjXzo3/pRZKfwBeLOkjeMIVIcydJjCiv0k3SvpJkm7\nZG1I0omS5kma199fZGQ857ItXZ498F7edOe6TWmVgpk9AHwduAX4BSF9dToP+t3ANma2G2E81Mzx\nk83sYjObZWazenvrpu5wLtcW06Y0NN25blNq7iMzuxS4FEDS1whDyyXnP5d4fqOk70jaxMyeLrNc\nrnPMXdDHeTcvYunyAbaYNoXZB88EGDatVe39sw+eOSSmADClZ2J1v851u1IrBUmbmtlTkrYGjgL2\nS83fDHjSzEzS3oRfLs+UWSbXObKCvrOvuRcMBtdYdVorA8GVbZRV6Tg31pWdJfVaSRsTBtY+2cye\nlXQSgJnNAY4GPiFpFTAAHGM+6k/XyAr6Dq4e/vZXAsGt+uA+co8ZXgk4l6Ps5qM3ZUybk3h+AXBB\nmWVwnauR4K4Hgp1rD+/R7EZNI8FdDwQ71x5jbpCd8S4r8DoemzrmLujjxZfTN6NBz0QNiSmAB4Kd\nayevFDpIt/S2TR9nxfSpPZx1eOiq0g0Vo3OdyCuFDlKrt+14+lDMOk6AqZMnVY9zPB2vc2OJxxQ6\nSLf0tu2W43RuLPJKoYN0S2/bbjlO58YirxQ6yOyDZzKlZ+KQaVlB1k5J/dxsOYoepytHp1w/rjN5\nTKGDFOlt2ynB6JGUw3sVj55OuX5c59JY60A8a9Ysmzdv3mgXY9Tsf+6t9GW0vc+YNoXfnfbWriuH\na4y/b91L0nwzm1VvOW8+GmM6JUjbKeVwjfH3zdXjlcIY0ylB2k4ph2uMv2+uHq8UWqzsIF6nBGmz\nytEzUbz48qpCx+7BztHRKdePa0w7/1880NxC7QjidUqQNl2OaVN7eOGlVSwfGARqH7sHO0dPp1w/\nrrh2/794oLmFujmI18ixd/N5cq5Rrfp/8UDzKOjmIF4jx97N58m5RrX7/8UrhRbq5iBeI8fezefJ\nuUa1+/+l1EpB0qcl/UHS/ZJOzZgvSedLekjSfZL2LLM8ZWtHEK+VAadGtlVv2UaOvdHz1MwxeyDb\njRftvjmgtECzpNcDHwf2Bl4BfiHpBjP7U2KxQ4Ad42Mf4ML4d0wqO4jXyoBTI9sqsmwjx97Iss0c\nswey3XjS7psDSgs0S3ofcLCZfSy+/iLwspl9I7HMRcDtZnZFfL0IOMDMnsjbbicHmsvWygDtWAkM\nN7NvD2Q7N1wnBJr/ALxZ0saSpgKHAlullpkBLE68XhKnDSHpREnzJM3r7+8vrcCdrpUBp7ESGG5m\n3x7Idq55pVUKZvYA8HXgFuAXwL1AevxFZa2asa2LzWyWmc3q7e1teVnHilYGnMZKYLiZfXsg27nm\nlRpoNrNLzWxPM3szsAz4U2qRJQz99bAlsLTMMo1lrQw4lRkYbqVm9u29dp1rXqk9miVtamZPSdoa\nOArYL7XI9cApkn5CCDCvqBVP6HatDDiVFRhutWb27b12nWteqT2aJf0G2BgYBP7BzP5b0kkAZjZH\nkoALgHcCK4ETzKxmFLmbA83OOdesooHmUn8pmNmbMqbNSTw34OQyy+Ccc64479HsnHOuyrOktsnc\nBX186Wf38+zKkEV02pQezn73Lrkdtoq2h89d0MfZ199fzU663uSJ9EycwIqBwbqdwop2Hiu7bT5r\nH+AxAedGg2dJbYO5C/qYfc29DK4eeq57Jojz3rdbzTGYIdw5c85Ru2b2NJ599b0Mrsl/D7PWLbqP\nRsrSrKx99EwUGEOOq9X7da7bdELnNRedd/OiYRUChA+9825eNGzZ5AckwMDg6mHLVbdbo0LIW7fo\nPhopS7Oy9jG42oYdV6v365zL5pVCGzTS+7YVPY1btY929AxuZFveI9m58nml0AaN9L5tRU/jVu2j\nHT2DG9mW90h2rnxeKbRIrVTNsw+eGdrJU3omiAN36h2y3oE79Q7rjStCps/M7U7IyhSyVlZP3qI9\nftvRMzhvrOf0cXmPZOfaw+8+aoF6qZorwdH03UeH7bY5187vG7LetfP7eO9eM7jtwX76lg8g1iaD\nyttuo3cfFe3x246ewXn7KHu/zrlsfvdRCzSbqrneep4C2jnXKn73URs1G5Ctt56ngHbOtZtXCi3Q\nbEC23nqeAto5124eU6ijXo/euQv6WPlKepiIYoHR2QfPzOzUduBOvcxd0MeLL2dv98Cdetn9S7+s\nxhGmT+3hrMOze0c751wjvFKooV4AOas3LtROYZG2OqNT2xV3LebKuxYP68A1fWoP73rD5sPmPbty\nkNnX3Fstl3PONcubj2qo16M3az7AeutMKvThfN7Ni1iTMX31muE9egGmTp7EbQ/2Z84bXD28d7Rz\nzjXKfynUUHYguNGAcbOBa+ecK8p/KdRQdiC40YDxFtOmNDU2sXPOFVX2cJyfAT5G6H+1kDCy2kuJ\n+ccD5wGVbroXmNklZZapEbMPnpmZJbTSCzmrDwGs7X1cr8PVgTv1ctkdjw+bPnGCmMDwLKGVwHVW\nZtSeicoMbJ8xdyFX3LmY1WZMlNh3++k8+sxAQ53C2pE+2znXGUqrFCTNAD4F7GxmA5KuAo4BfpBa\n9EozO6WscoxEVm/bA3fqHdILOU86KJ02d0Ef187vGzZ9vckT+ep7dh223/QHcbIXc97dR2fMXTik\n0lltxu8eXla4jJVy1gq2O+fGl7JjCpOAKZIGganA0pL313LJdBIQeiHXqxAqKkHprA/PvCD1tKmT\nq8vnfeimy5TnijsXj6iMeeWst45zbuwqLaZgZn3AN4HHgSeAFWb2y4xF3yvpPknXSNoqa1uSTpQ0\nT9K8/v7+sopcSKuCw+3orby6YAqTRlJ7F1nHOTd2lVYpSJoOHAFsB2wBrCfpuNRiPwO2NbM3AP8F\n/DBrW2Z2sZnNMrNZvb29ZRW5kGaCw62Y3oyJqp1Btcg+vVe1c92lzLuP3gb82cz6zWwQuA746+QC\nZvaMmb0cX34X2KvE8rREVqrnPHnB38p20umhJwhefHnVsPTbZ8xdyGtOv5FtT7uB15x+I2fMXVho\n/8fuk/nDa4h6Pa/bkT7bOdc5yowpPA7sK2kqMAAcBAxJbyppczN7Ir58N/BAieVpiWTwOe/uo6p6\nrTepL/JrjGrwuBLQvXre40OCw6vNqsHjrxy5a83NV+aP5O6jdqTPds51jlJTZ0v6EvABYBWwgHB7\n6heAeWZ2vaRzCJXBKmAZ8Akze7DWNjspdXat21Ir8tJcF1m3lokSD59zaNPrO+e6S9HU2aXefWRm\nZwFnpSafmZh/OnB6mWUoU5Fga1mB2qJBZOeca4T3aB6BIsHWsgK1RYPIzjnXCK8URuDAnXrTYYEh\nagVkiwSseyaI/V+zUea8dBC51hjRjWrltkbTeDkO59rJK4UmVXokpxtxpvRMQIRYwjlH7VqzA9p7\n95pRs1JB8L5ZW3PcvltXfxlMlDhu362HBJkrvY77lg9grA1SN/Mh2MptjabxchzOtZtnSW1SXo/k\njdZbp/D4ybc92F/zBqVKOuzfnfbWmncatbLX8XjpwTxejsO5dvNfCk1qRQB5JIHqVpeljG2NpvFy\nHM61m1cKTWpFAHkkgepWl6WMbY2m8XIczrWbVwpNakVP33rB5qLby9qOWJvCe+6CvsJB17J6MLc7\n6Os9sZ1rjscUmtSKnr7pbWw4pQcJlq8cbGh76V7WYm1n6r7lA2H8Zls7PkOt9Ndl9GAejfTb3hPb\nueaU2qO5DJ3Uo7kTNdJTOq+3davllald+3fOFe/R7M1H40yrA92t4EFf58YOrxTGmVYHulvBg77O\njR0eU2hAerzjY/fZqm6m0oqscY6h9W3eWeNK90zUkJgCtDfomjfWtQd9nes8XikUlDXecdEU1lmB\n1tlX3wsKHdQq01oRfM0LsGZNa1fQ1YO+zo0dHmgu6DWn35iZmbRICutODP4657qLB5pbLC9VdZEU\n1p0Y/HXOuSxeKRSUl6q6SArrTgz+OudcllJjCpI+QxhtzYCFwAlm9lJi/jrAjwhjMz8DfMDMHi2z\nTBVZgd90G3dymUkTYHXGj4J9t5/O/ufeWrPzWWbwd4KGxBTAg6/OudFXWqUgaQbwKWBnMxuQdBVw\nDPCDxGJ/CzxrZjtIOgb4OmH4zlIV6WGbXmZwzfDt7Ljpetz9+IrqMpXxlfO22UnBX+ecy1L23UeT\ngCmSBoGpwNLU/COAs+Pza4ALJMlKjn4XSauclxo76ZH+lTVjCsltVh5pXgk45zpJaTEFM+sDvgk8\nDjwBrDCzX6YWmwEsjsuvAlYAG6e3JelESfMkzevv7x9x2Yr0sC0S8G11kNk550ZbaZWCpOmEXwLb\nAVsA60k6Lr1YxqrDPmnN7GIzm2Vms3p7e0dctiI9bIsEfFsdZHbOudFWZvPR24A/m1k/gKTrgL8G\nLkssswTYClgiaRKwIbCsxDIBxXrYHrhTL5ff8XjuyGgi/FJIZiRNq2xz7oI+zr7+/mrMYfrUHt71\nhs257cH+uvGEIgFx55xrlTIrhceBfSVNBQaAg4B0r7PrgY8A/wMcDdxadjwB6vewrTX+8sDgmiEV\ngUH19bSMu48AZl9975AUE8+uHBzSOzqvN/NopJx2znW3QpWCpH3N7I5GNmxmd0q6BrgbWAUsAC6W\n9GVgnpldD1wK/FjSQ4RfCMc0VPoRyAv8Qu3xl4FhvZON/J7I+59765AKIU/W+ME+zrBzrt2K/lL4\nDrBnoxs3s7OAs1KTz0zMfwl4X6PbLVszqZ7LGLPZU04759qtaKC5fkR1HKkViG40DfRIejN7ymnn\nXLsV/aWwoaSj0hPN7LoWl2dUVIK5fcsHmChl3mqaDEQ3kgb6wJ16h8QP8mRtw1NOO+farWil8Gvg\n8NQ0A8Z8pZAO5mZVCALeu9eMYe39Re4cunb+8AHqp/ZM4Ki9tqx795GnnHbOtVvXp84umta6mZTW\nPjaxc65TFE2dXfTuo++T3am2EkHCAAAVVElEQVTso02UraMUDdo2E9z1QLFzbqwp2nz08/j3G8Dn\nSirLqNhi2pRCvxSaCe7mbdsDxc65TlXo7iMzu9bMrgWeqzyPr8e82QfPZErPxJrLNBvczdq2B4qd\nc52s0R7NYysAUcCRe8xg3mPLuOLOxUOCzJW7kGZk9HZuZByGdXsmMEGwxsI20wHrPMltTJvagxms\nGBhsabDZU2g459KKxhQWEiqEHSTdR8zsYGZvKLNw7VC5QyhZIUzpmcg5R+2a+WHf6DgMA4mBGFab\nce38PmZts1HND9/0Np5dWXuchmaP21NoOOfSinZeO4xwS+rr4t/K6zGvViqJZpatNw5D3rbr7afR\nbdTTyHE757pH0eaj50stxShq5A6hVo3DUG+ZVmyj2fX9zijnulvRXwpPA/cTspzOj4/WdRYYRY2k\nkmjVOAz1lmnFNppd3++Mcq67Fa0UTiSMffAtYEcz287Mti+vWK01d0Ef+597K9uddgP7n3srcxes\n7WWcd4fQgTv1Dltn9sEz6Zk4NA1Uz0RV7yaau6CPF19eVbMsRe4+qndHVCvuYOrmO6NqXQ/Odbui\nt6ReArwRWAf4vaQPlVqqFqoEVPuWD2CsDahWPgiO3GMG5xy1KzOmTUGE3sbv3WsG187vG7bOvMeW\nDb//yobupzKQTpasdBlZ0mWaPrUnjNUQy5cVBG9U1nG3Yrudrt714Fy3K5TmIpUMb0PgM4S7j3Yr\nq2B5Gk1z0Uyqibx18pLlzYhNLmWly3Ct46lHXLdqaZoLht9pNL/xIo2OVo6NkFUh1NvWSJZ1recB\ndudqK1opfNvM7i61JCVpJtVE3jp5vxS2aOCXggdyR5enHnGutqKB5ksa3bCkmZLuSTyek3RqapkD\nJK1ILHNm3vaa1UxA9cCdejNHFZo8ScMCzZVtNZMuY6wGPMdquaG7A+zOFVH0l8IkSdNJjcBmZsvy\nVjCzRcDuAJImAn3ATzMW/Y2ZHVawHA1rdEyCSg/nrIaigcE19EwQ06f2sHzl8JQT8x5bxuV3PD5k\nXbF2DOd0uoyx2KN4rJa7wseocK62opXCTEIcIVkpGFD0ttSDgIfN7LEGytYyR+5RLN8Q1O9NPLjG\nmDp5EgvOfMewebc92J95c1JWELNWj+JO/oAaq+VOauR6cK7bFK0U/mhme4xgP8cAV+TM20/SvcBS\n4LNmdn96AUknEvpKsPXWW4+gGPWNpDdxq3tHd6KxWm7nXDFFYwpNkzQZeDdwdcbsu4Ft4q2t3wbm\nZm3DzC42s1lmNqu3t7e8wjKy3sSt7h3dicZquZ1zxRStFPYbwT4OAe42syfTM8zsOTN7IT6/EeiR\ntMkI9jVi9QLGIgSiK5JB15WvrKJnQqrH8wSx8pVVw4KyzQY8RxLkbUWA2AO1zo1vRZuPrpKG349j\nZu8usO6x5DQdSdoMeNLMTNLehErqmYJlKkU6ELluz4Qh6a8NqumvgWEprnsmimlTelgxMMiGU3p4\n8ZVV1dTXWUHZRgKeIwnytipA7IFa58a3mj2aJZ1rZqdJ+g2wPvA1oPqN38x+VXPj0lRgMbC9ma2I\n006K686RdArwCWAVMAD8g5n9vtY2G+3RPFK1esBCdt+ESmC51b1nR7I978nrXHdrVY/mtwKY2Zsk\nvQv4PHAb8A0ze67exs1sJbBxatqcxPMLgAvqbWc0jaRHdKuDsiPZngeInXNF1Isp/KXyxMxuMLP9\nCSm0b5H02VJL1iFqBVbrBV1bHZQdyfY8QOycK6JepfA+AEnPxx7JzwEXA68Hvl524TpBrcBqMuBc\nkUyl3eqg7Ei25wFi51wRNZuPzOzl+Hf99hSn8+QFVgGuvGvxsOVXr7G66zYblB3J9jxA7Jwromjq\nbAEfArYzs3+StDWwmZndVXYB09odaM6TF7gFD9465zpP0UBz0X4K3yH0VfhgfP088O9Nlm1caCb1\ntnPOdbqilcI+ZnYy8BKAmT0LTC6tVGNAvdTbzjk3FhXtvDYYM50agKReYE3tVca+uQv6qm3wG07p\nQaKaHfXAnXq58q7FDK4Z2vyWDDQ759xYU/SXwvmEtNebSvoq8FtCR7ZxKz2W7/KBQZ5dOVgd1/fa\n+X18YO+tmDalp7rO9Kk9nHf0bh68dc6NWYV+KZjZ5ZLmE1JgCzjSzB4otWSjrF4K7YHB1dz2YD/3\nnDU8hbZzzo1VNSsFSesCJwE7AAuBi8xsVTsKNtpG0kvYOefGqnrNRz8EZhEqhEOAb5Zeog4xkl7C\nzjk3VtVrPtrZzHYFkHQp0PZ+CWWqBJL7lg8wUWK1WXXYzNkHzxySVTTNewM758ajer8UBitPxluz\nUTKQDLA6duJLppQ+56hdmTFtCgKmTelh+tQeROicds5Ru3pA2Tk37tT7pbBbzHcEIcA8Jb4WYGa2\nQamlK1GtQHJlzOHfnfZW/+B3znWVermP8ocgG+PqBYk9iOyc60alj9HcqeoFiT2I7JzrRkV7NDdM\n0kzgysSk7YEzzexfE8sI+DfgUGAlcLyZ3V1GeZK9kys9kq+d35fZhOS9kp1z3aq0XwpmtsjMdjez\n3YG9CB/6P00tdgiwY3ycCFxYRlnSvZMrPZLfu9cMpk/tGb5C/cSxzjk3LrWr+egg4GEzeyw1/Qjg\nRxbcAUyTtHmrd54VVK70SJ46efiPpcE1xnk3L2p1MZxzruOV1nyUcgxwRcb0GUBypJolcdoTyYUk\nnUj4JcHWW2/d8M5HMs6yc851k9J/KUiaDLwbuDprdsa0YY03Znaxmc0ys1m9vcOHwKxnJOMsO+dc\nN2lH89EhwN1m9mTGvCXAVonXWwJLW12AWuMT+9jF9c1d0Mf+597KdqfdwP7n3srcBX2jXSTnXEna\n0Xx0LNlNRwDXA6dI+gmwD7DCzJ7IWbZpRcYn9rGLs1WC9JWYTLLHt58j58afQmM0N71xaSohZrC9\nma2I004CMLM58ZbUC4B3Eu5OOsHMag7A3CljNHeLvLGofRxq58aWomM0l/pLwcxWAhunps1JPDfg\n5DLL4EammSC9c27s6toeza4YD8Q7113adUtqx0r3dB5JPKGV2+oUWSnEPRDv3PjV1ZVCK4Oo4zUg\nWyRI75wbP7q6Usjr6XzezYsa/tBr5bY6zZF7zBjzx+CcK6arYwqtDKJ6QNY5Nx50daXQyiCqB2Sd\nc+NBV1cKrezN7D2jnXPjQVfHFFoZRPWArHNuPCi1R3MZvEezc841rmiP5q5uPnLOOTeUVwrOOeeq\nujqmMBaMx17SzrnO5ZVCBxuvvaSdc53Lm486WK1e0s45VwavFDqY95J2zrWbVwodzHtJO+fardRK\nQdI0SddIelDSA5L2S80/QNIKSffEx5lllqfT1Bv72HtJO+farexA878BvzCzoyVNBqZmLPMbMzus\n5HJ0nCJBZO8l7Zxrt9IqBUkbAG8Gjgcws1eAV8ra31hTNNW2p612zrVTmc1H2wP9wPclLZB0iaT1\nMpbbT9K9km6StEvWhiSdKGmepHn9/f0lFrl9PIjsnOtEZVYKk4A9gQvNbA/gReC01DJ3A9uY2W7A\nt4G5WRsys4vNbJaZzert7S2xyO3jQWTnXCcqs1JYAiwxszvj62sIlUSVmT1nZi/E5zcCPZI2KbFM\nueoFfVvNg8jOuU5UWkzBzP4iabGkmWa2CDgI+GNyGUmbAU+amUnam1BJPVNWmfKMRs9hDyI75zpR\n2XcffRK4PN559AhwgqSTAMxsDnA08AlJq4AB4BgbhVzeozW+sgeRnXOdptRKwczuAdL5u+ck5l8A\nXFBmGYrwoK9zzgXeoxkP+jrnXEVXZ0mtpKXuWz6AgGS7lQd9nXPdqGsrhXRw2aBaMczwoK9zrkt1\nbaWQFVyuVAi/O+2to1Mo55wbZV0bU/DgsnPODde1lYIHl51zbriurRS8R7Fzzg3XtTEF71HsnHPD\ndW2lAN6j2Dnn0rq2+cg559xwXik455yr8krBOedclVcKzjnnqrxScM45V+WVgnPOuSqvFJxzzlWV\nWilImibpGkkPSnpA0n6p+ZJ0vqSHJN0nac+8bTnnnCtf2Z3X/g34hZkdHYfknJqafwiwY3zsA1wY\n/zrnnBsFpf1SkLQB8GbgUgAze8XMlqcWOwL4kQV3ANMkbV5WmZxzztVWZvPR9kA/8H1JCyRdImm9\n1DIzgMWJ10vitCEknShpnqR5/f395ZXYOee6XJmVwiRgT+BCM9sDeBE4LbWMMtazYRPMLjazWWY2\nq7e3t/Uldc45B5RbKSwBlpjZnfH1NYRKIr3MVonXWwJLSyyTc865GkqrFMzsL8BiSZUBCg4C/pha\n7Hrgw/EupH2BFWb2RFllcs45V1vZdx99Erg83nn0CHCCpJMAzGwOcCNwKPAQsBI4oeTyOOecq6HU\nSsHM7gFmpSbPScw34OQyy+Ccc64479HsnHOuyisF55xzVV4pOOecq/JKwTnnXJVXCs4556q8UnDO\nOVdVdj+FrjJ3QR/n3byIpcsH2GLaFGYfPJMj9xiWysk55zqWVwotMndBH6dft5CBwdUA9C0f4PTr\nFgJ4xeCcGzO8+ahFzrt5UbVCqBgYXM15Ny8apRI551zjvFJokaXLBxqa7pxzncgrhRbZYtqUhqY7\n51wn8kqhRWYfPJMpPROHTJvSM5HZB8/MWSOYu6CP/c+9le1Ou4H9z72VuQv6yiymc87V5IHmFqkE\nkxu5+8iD0865TuOVQgsduceMhj7MawWnvVJwzo0Gbz4aRR6cds51Gq8URpEHp51znabUSkHSo5IW\nSrpH0ryM+QdIWhHn3yPpzDLL02maDU4751xZ2hFTONDMnq4x/zdmdlgbytFxmglOO+dcmTzQPMoa\nDU4751yZyo4pGPBLSfMlnZizzH6S7pV0k6RdshaQdKKkeZLm9ff3l1da55zrcmX/UtjfzJZK2hS4\nRdKDZvbrxPy7gW3M7AVJhwJzgR3TGzGzi4GLAWbNmmUll9k557pWqb8UzGxp/PsU8FNg79T858zs\nhfj8RqBH0iZllsk551y+0ioFSetJWr/yHHgH8IfUMptJUny+dyzPM2WVyTnnXG1lNh+9Gvhp/Myf\nBPyHmf1C0kkAZjYHOBr4hKRVwABwjJl585Bzzo2S0ioFM3sE2C1j+pzE8wuAC8oqg3POucZorH0x\nl9QPPDba5WjSJkCtPhvdxs/HUH4+hvLzMdRIz8c2ZtZbb6ExVymMZZLmmdms0S5Hp/DzMZSfj6H8\nfAzVrvPhuY+cc85VeaXgnHOuyiuF9rp4tAvQYfx8DOXnYyg/H0O15Xx4TME551yV/1JwzjlX5ZWC\nc865Kq8UmpA1eJCkjSTdIulP8e/0OF2Szpf0kKT7JO2Z2M5H4vJ/kvSRxPS94vYfiuuq/UeZT9L3\nJD0l6Q+JaaUff94+RlvO+ThbUl9iAKlDE/NOj8e2SNLBienvjNMeknRaYvp2ku6Mx32lpMlx+jrx\n9UNx/rbtOeLaJG0l6TZJD0i6X9Kn4/SuvEZqnI/OvEbMzB8NPoBHgU1S074BnBafnwZ8PT4/FLgJ\nELAvcGecvhHwSPw7PT6fHufdBewX17kJOGS0jzl1rG8G9gT+0M7jz9vHaD9yzsfZwGczlt0ZuBdY\nB9gOeBiYGB8PA9sDk+MyO8d1riKkgAGYA3wiPv97YE58fgxw5Wifi1iWzYE94/P1gf+Lx92V10iN\n89GR18ioX0Bj8UF2pbAI2DxxESyKzy8Cjk0vBxwLXJSYflGctjnwYGL6kOU65QFsy9APwdKPP28f\nnfDIOB95//CnA6cnXt8cP9z2A25OLxc/9J4GJsXp1eUq68bnk+JyGu1zkXHM/wm8vduvkYzz0ZHX\niDcfNSdr8KBXm9kTAPHvpnH6DGBxYt0lcVqt6Usypne6dhx/3j461SmxOeR7iWaMRs/HxsByM1uV\nmj5kW3H+irh8x4jNFXsAd+LXSPp8QAdeI14pNGd/M9sTOAQ4WdKbayybFQ+wJqaPVd16/BcCrwF2\nB54AvhWnt/J8dPS5kvQq4FrgVDN7rtaiGdPG3TWScT468hrxSqEJlj140JOSNgeIf5+Kiy8Btkqs\nviWwtM70LTOmd7p2HH/ePjqOmT1pZqvNbA3wXdYOMNXo+XgamCZpUmr6kG3F+RsCy1p/NI2T1EP4\nALzczK6Lk7v2Gsk6H516jXil0CDlDx50PVC5O+IjhHZD4vQPxzss9gVWxJ+1NwPvkDQ9/mx8B6Ed\n8AngeUn7xjsqPpzYVidrx/Hn7aPjVD6YovewdoCp64Fj4l0h2xGGn70L+F9gx3gXyWRCUPB6C43B\ntxHGHoHh57ZyPo4Gbo3Lj6r4vl0KPGBm/5yY1ZXXSN756NhrZLSDLmPtQYj83xsf9wNfiNM3Bv4b\n+FP8u1GcLuDfCXcNLARmJbb1UeCh+DghMX1WvEAeJow30VHBQ+AKws/dQcI3kb9tx/Hn7WO0Hznn\n48fxeO+L/5ibJ5b/Qjy2RSTuLCPchfN/cd4XUtfcXfE8XQ2sE6evG18/FOdvP9rnIpbrjYQmivuA\ne+Lj0G69Rmqcj468RjzNhXPOuSpvPnLOOVfllYJzzrkqrxScc85VeaXgnHOuyisF55xzVV4puHFP\n0maSfiLpYUl/lHSjpNeOdrmc60ReKbhxLXYc+ilwu5m9xsx2Bj4PvHp0S+ZcZ/JKwY13BwKDZjan\nMsHM7gFOijnsl0n6c3x+kqR1JX1fIVf/AkkHAkg6XtIF8flMSaskHR1fPyrpisr2Y/76R+PzvO1N\nlPTNOP0+SZ+U9IFYjockrYjPb4zLv9Cm8+W63KT6izg3pr0emJ+eaGYfApD0A+DnZnZNfP2Pcf6u\nknYiZMNNNzX9E/BgatoWMRWDgM0S00/O2d4JhFz5e5jZKkkbmdky4EpJBxBSKh82guN2rin+S8G5\nod5ISD+AmT0IPAZUKwVJexH+b+al1rsC+GB8/EeB7b2NMPjJqjivXiK7KfGXw70KI435/64rhV9Y\nbry7H9irgeXrDX36FeCLGdOvB94dHz8rsD3RWLrnATPbnXAsbyBUKs61nFcKbry7FVhH0scrEyT9\nlaS35Cz/a6DStPRaYGtCUjKAtwBPmNkDGeu9AtwB/E98Xm97vyTENSbFeRsVORhbO1DK5CLLO9co\nrxTcuGYh4+N7gLfHW1LvJwyDmDdGxXeAiZIWAlcCx5vZy3HejnHdvH2dZWZnFdzeJcDjwH2S7iU0\nO9UyRdJvJd0JvERIK+1cy3mWVOecc1X+S8E551yVVwrOOeeqvFJwzjlX5ZWCc865Kq8UnHPOVXml\n4JxzrsorBeecc1X/H7PMvQC4wmPuAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x24965e1f9e8>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(data.price, data.rating)\n",
    "plt.xlabel('Стоимость')\n",
    "plt.ylabel('Рейтинг')\n",
    "plt.title('Как и следовало ожидать, \\n у более дорогих отелей рейтинг в среднем выше');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Теперь найдем Парето-оптимальные отели.\n",
    "\n",
    "Напомним, что отель называется Парето-оптимальным, если никакой другой отель его не доминирует. \n",
    "\n",
    "Мы говорим, что отель A доминирует Б, если A не хуже Б ни по цене, ни по качеству, и хотя бы по одному из параметров А строго лучше. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>name</th>\n",
       "      <th>price</th>\n",
       "      <th>rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>83</th>\n",
       "      <td>Jingtailong International Hotel</td>\n",
       "      <td>26985.0</td>\n",
       "      <td>7.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>Beijing Prime Hotel Wangfujing</td>\n",
       "      <td>30160.0</td>\n",
       "      <td>8.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>93</th>\n",
       "      <td>Beijing Qianyuan Hotel</td>\n",
       "      <td>37887.0</td>\n",
       "      <td>8.6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Beijing Hotel NUO Wangfujing</td>\n",
       "      <td>45401.0</td>\n",
       "      <td>8.7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>New World Beijing Hotel</td>\n",
       "      <td>46554.0</td>\n",
       "      <td>9.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>Shichahai Sandalwood Boutique Hotel Beijing</td>\n",
       "      <td>58181.0</td>\n",
       "      <td>9.3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>99</th>\n",
       "      <td>Zhaolin Grand Hotel</td>\n",
       "      <td>102196.0</td>\n",
       "      <td>9.4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57</th>\n",
       "      <td>Rosewood Beijing</td>\n",
       "      <td>123717.0</td>\n",
       "      <td>9.5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                           name     price  rating\n",
       "83              Jingtailong International Hotel   26985.0     7.6\n",
       "18               Beijing Prime Hotel Wangfujing   30160.0     8.0\n",
       "93                       Beijing Qianyuan Hotel   37887.0     8.6\n",
       "17                 Beijing Hotel NUO Wangfujing   45401.0     8.7\n",
       "0                       New World Beijing Hotel   46554.0     9.1\n",
       "52  Shichahai Sandalwood Boutique Hotel Beijing   58181.0     9.3\n",
       "99                          Zhaolin Grand Hotel  102196.0     9.4\n",
       "57                             Rosewood Beijing  123717.0     9.5"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "optimal = []\n",
    "# проходим по списку всех отелей по возрастанию цены\n",
    "for idx, hotel in data.sort_values('price').iterrows():\n",
    "    # берём все отели, которые не хуже данного по цене и качеству\n",
    "    better = data[(data.price <= hotel.price) & (data.rating >= hotel.rating)]\n",
    "    # если ни один из них не строго лучше нашего, то наш - Парето-оптимальный\n",
    "    if better.price.min() == hotel.price and better.rating.max() == hotel.rating:\n",
    "        optimal.append(hotel)\n",
    "optimal = pd.DataFrame(optimal)\n",
    "optimal"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Именно эти 8 отелей обеспечивают лучшее соотношение цена-качество. \n",
    "\n",
    "Все остальные отели хуже какого-либо из Парето-оптимальных или по цене, или по качеству, или по тому и другому. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYUAAAElCAYAAAALP/6mAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nCGsUjDHGRPx/2pexHBh2t1EAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x249681b9f60>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(data.price, data.rating, s=5);\n",
    "plt.plot(optimal.price, optimal.rating, color = 'red')\n",
    "plt.xlabel('Стоимость')\n",
    "plt.ylabel('Рейтинг')\n",
    "plt.title('Парето-граница - это 8 отелей \\n с наилучшим соотношением цена/качество');\n",
    "plt.text(100000, 7, 'А нашли мы их, распарсив booking.com!');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Напоследок сохраним данные, чтобы этот блокнот стал воспроизводим. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import pickle\n",
    "pickle.dump(hotels, open('hotels.pkl', 'wb'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Удачного вам парсинга, оптимизации, и, конечно, новогодних праздников! До встречи :)"
   ]
  }
 ],
 "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": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}