{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img align=\"center\" src=\"https://raw.githubusercontent.com/FUlyankin/Parsers/master/images%20/cats.jpg\" height=\"1200\" width=\"1200\"> \n",
    "\n",
    "# Домашка №4: визуализация\n",
    "\n",
    "У всех нас есть датасет по контакту. В нём лежит информация про всех нас. Эту информацию надо как следует проанализировать. Именно этим мы сейчас продолжим заниматься. В этот раз с картинками! \n",
    "\n",
    "Для начала подгрузите все необходимые библиотеки: `pandas`, `matplotlib.pyplot`, `seaborn` и включите опцию, отвечающую за прорисовку картинок прямо в питонячей тетрадке."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Подгрузите данные профилей и данные по фотографиям"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# В этой табличке лежат данные по профилю человека\n",
    "df_profile = pd.read_csv('../data/vk_download/vk_data_profile.csv',sep='\\t')\n",
    "\n",
    "# В этой табличке лежат данные по фотографиям человека\n",
    "df_photo = pd.read_csv('../data/vk_download/vk_data_photo.csv',sep='\\t')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Склеим табличку по полю uid"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0_x</th>\n",
       "      <th>uid</th>\n",
       "      <th>Unnamed: 0.1</th>\n",
       "      <th>first_name</th>\n",
       "      <th>last_name</th>\n",
       "      <th>is_closed</th>\n",
       "      <th>city</th>\n",
       "      <th>home_town</th>\n",
       "      <th>male_dummy</th>\n",
       "      <th>relation_cat</th>\n",
       "      <th>...</th>\n",
       "      <th>photo_repost_cnt</th>\n",
       "      <th>photo_repost_max</th>\n",
       "      <th>photo_repost_mean</th>\n",
       "      <th>photo_repost_median</th>\n",
       "      <th>photo_text</th>\n",
       "      <th>photo_text_len_cnt</th>\n",
       "      <th>photo_yer_mean</th>\n",
       "      <th>vk_photo_ava_change_cnt</th>\n",
       "      <th>vk_photo_text_url_len_cnt</th>\n",
       "      <th>vk_photo_wall_ph_post_cnt</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>182152789</td>\n",
       "      <td>0</td>\n",
       "      <td>Александра</td>\n",
       "      <td>Абашкова</td>\n",
       "      <td>False</td>\n",
       "      <td>Москва</td>\n",
       "      <td>Москва</td>\n",
       "      <td>0</td>\n",
       "      <td>не указано</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td></td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.333333</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>148020433</td>\n",
       "      <td>1</td>\n",
       "      <td>Анастасия</td>\n",
       "      <td>Чуфистова</td>\n",
       "      <td>False</td>\n",
       "      <td>Рязань</td>\n",
       "      <td>Рязань</td>\n",
       "      <td>0</td>\n",
       "      <td>не указано</td>\n",
       "      <td>...</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.105263</td>\n",
       "      <td>0.0</td>\n",
       "      <td></td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.166667</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>138413935</td>\n",
       "      <td>2</td>\n",
       "      <td>Александр</td>\n",
       "      <td>Головачев</td>\n",
       "      <td>False</td>\n",
       "      <td>Москва</td>\n",
       "      <td>Омск</td>\n",
       "      <td>1</td>\n",
       "      <td>не женат/не замужем</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td></td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.333333</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>366261055</td>\n",
       "      <td>3</td>\n",
       "      <td>Анна</td>\n",
       "      <td>Лобанова</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>не указано</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td></td>\n",
       "      <td>0.0</td>\n",
       "      <td>12.500000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>111252392</td>\n",
       "      <td>4</td>\n",
       "      <td>Алексей</td>\n",
       "      <td>Пузырный</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td></td>\n",
       "      <td>0.0</td>\n",
       "      <td>8.750000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 85 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0_x        uid  Unnamed: 0.1  first_name  last_name  is_closed  \\\n",
       "0             0  182152789             0  Александра   Абашкова      False   \n",
       "1             1  148020433             1   Анастасия  Чуфистова      False   \n",
       "2             2  138413935             2   Александр  Головачев      False   \n",
       "3             3  366261055             3        Анна   Лобанова      False   \n",
       "4             4  111252392             4     Алексей   Пузырный      False   \n",
       "\n",
       "     city home_town  male_dummy         relation_cat  ... photo_repost_cnt  \\\n",
       "0  Москва    Москва           0           не указано  ...              0.0   \n",
       "1  Рязань    Рязань           0           не указано  ...              2.0   \n",
       "2  Москва      Омск           1  не женат/не замужем  ...              0.0   \n",
       "3     NaN       NaN           0           не указано  ...              0.0   \n",
       "4     NaN       NaN           1                  NaN  ...              0.0   \n",
       "\n",
       "   photo_repost_max  photo_repost_mean  photo_repost_median  \\\n",
       "0               0.0           0.000000                  0.0   \n",
       "1               1.0           0.105263                  0.0   \n",
       "2               0.0           0.000000                  0.0   \n",
       "3               0.0           0.000000                  0.0   \n",
       "4               0.0           0.000000                  0.0   \n",
       "\n",
       "                                          photo_text  photo_text_len_cnt  \\\n",
       "0                                                                    0.0   \n",
       "1                                                                    0.0   \n",
       "2                                                                    0.0   \n",
       "3                                                                    0.0   \n",
       "4                                                                    0.0   \n",
       "\n",
       "   photo_yer_mean  vk_photo_ava_change_cnt  vk_photo_text_url_len_cnt  \\\n",
       "0        1.333333                      0.0                        0.0   \n",
       "1        3.166667                      0.0                        0.0   \n",
       "2        2.333333                      0.0                        0.0   \n",
       "3       12.500000                      0.0                        0.0   \n",
       "4        8.750000                      0.0                        0.0   \n",
       "\n",
       "   vk_photo_wall_ph_post_cnt  \n",
       "0                        0.0  \n",
       "1                        0.0  \n",
       "2                        0.0  \n",
       "3                        0.0  \n",
       "4                        0.0  \n",
       "\n",
       "[5 rows x 85 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.merge(df_profile, df_photo, how='right', on='uid')\n",
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Давайте посмотрим на все названия колонок, которые есть в таблице. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['Unnamed: 0_x', 'uid', 'Unnamed: 0.1', 'first_name', 'last_name',\n",
       "       'is_closed', 'city', 'home_town', 'male_dummy', 'relation_cat',\n",
       "       'relation_partner', 'byear', 'bmonth', 'bday', 'country',\n",
       "       'facebook_dummy', 'instagram_dummy', 'skype_dummy', 'twitter_dummy',\n",
       "       'home_phone_dummy', 'mobile_phone_dummy', 'site_dummy', 'folowers_cnt',\n",
       "       'university_str', 'faculty_str', 'about_str', 'activities_str',\n",
       "       'books_str', 'interests_str', 'movies_str', 'music_str', 'quotes_str',\n",
       "       'tv_str', 'games_str', 'can_post_dummy', 'can_see_all_posts_dummy',\n",
       "       'can_see_audio_dummy', 'can_write_private_message_dummy',\n",
       "       'has_mobile_dummy', 'has_ava_dummy', 'wall_comments_dummy',\n",
       "       'albums_cnt', 'audio_cnt', 'followers_cnt', 'friends_cnt', 'gifts_cnt',\n",
       "       'groups_cnt', 'mutual_friends_cnt', 'photos_cnt', 'subscriptions_cnt',\n",
       "       'user_photos_cnt', 'videos_cnt', 'pages_cnt', 'alco_love_cat',\n",
       "       'smoke_love_cat', 'religion_str', 'inspired_by_str', 'life_main_cat',\n",
       "       'people_main_cat', 'political_cat', 'lang_cnt', 'english_dummy',\n",
       "       'change_city_school_cnt', 'last_bukva_class_str', 'schools_cnt',\n",
       "       'is_bmm', 'in_hse_memes_group', 'likes_memes', 'Unnamed: 0_y',\n",
       "       'photo_cnt', 'photo_like_cnt', 'photo_like_max', 'photo_like_mean',\n",
       "       'photo_like_median', 'photo_month_mean', 'photo_repost_cnt',\n",
       "       'photo_repost_max', 'photo_repost_mean', 'photo_repost_median',\n",
       "       'photo_text', 'photo_text_len_cnt', 'photo_yer_mean',\n",
       "       'vk_photo_ava_change_cnt', 'vk_photo_text_url_len_cnt',\n",
       "       'vk_photo_wall_ph_post_cnt'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Мы видим, что часть из них кончается на `cnt`. Это колонки-счётчики. В них лежат такие переменные как количество фоток, лайков, репостов и тд"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['folowers_cnt',\n",
       " 'albums_cnt',\n",
       " 'audio_cnt',\n",
       " 'followers_cnt',\n",
       " 'friends_cnt',\n",
       " 'gifts_cnt',\n",
       " 'groups_cnt',\n",
       " 'mutual_friends_cnt',\n",
       " 'photos_cnt',\n",
       " 'subscriptions_cnt',\n",
       " 'user_photos_cnt',\n",
       " 'videos_cnt',\n",
       " 'pages_cnt',\n",
       " 'lang_cnt',\n",
       " 'change_city_school_cnt',\n",
       " 'schools_cnt',\n",
       " 'photo_cnt',\n",
       " 'photo_like_cnt',\n",
       " 'photo_repost_cnt',\n",
       " 'photo_text_len_cnt',\n",
       " 'vk_photo_ava_change_cnt',\n",
       " 'vk_photo_text_url_len_cnt',\n",
       " 'vk_photo_wall_ph_post_cnt']"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "variables_cnt = [item for item in df.columns if item[-3:] == 'cnt']\n",
    "variables_cnt"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Разбиритесь как работает и устроен этот цикл, если вы ещё не разбираетесь. Почитать об этом можно [вот тут.](https://habr.com/ru/post/30232/) Чуть ниже вам предстоит написать свой такой. \n",
    "\n",
    "Вытащите из переменных счётчиков только те, которые отвечают за фото. Постройте для них гистограммы (для удобства используйте логарифмическое скалирование). Как думаете, в каких переменных есть выбросы? Какие из переменных неинформативны? Почему?\n",
    "\n",
    "__Ответ:__ Выбросы есть везде, где длинные хвосты. То есть это почти каждая переменная. Неинформативная `photo_ava_change_cnt`. Она принимает только одно значение. Какой вообще в ней смысл?!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "photo_vars = [var_name for var_name in variables_cnt if 'photo' in var_name]\n",
    "df[photo_vars].hist(figsize = (15,8), log=True);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Кто больше постит фотографий, девушки или парни? чьи фото собирают больше лайков? (общее количество лайков, медиана и среднее)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/site-packages/numpy/lib/histograms.py:839: RuntimeWarning: invalid value encountered in greater_equal\n",
      "  keep = (tmp_a >= first_edge)\n",
      "/usr/local/lib/python3.7/site-packages/numpy/lib/histograms.py:840: RuntimeWarning: invalid value encountered in less_equal\n",
      "  keep &= (tmp_a <= last_edge)\n",
      "/usr/local/lib/python3.7/site-packages/matplotlib/figure.py:445: UserWarning: Matplotlib is currently using module://ipykernel.pylab.backend_inline, which is a non-GUI backend, so cannot show the figure.\n",
      "  % get_backend())\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAskAAAVICAYAAABWfkceAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzs3X24ZmV9H/rvj9cJihuBNFUGnDGDvF1osFOU0KaKngDRkRyv2kK4GlHjHI6a0zZRD6jTztFqTUrbcyBarqEQisV3lJdKKtJCjY3KSxCDQcJARhiighB2YFJkJt7nj73APTczzN4zz55nz+zP57qea6/nftbLb631rD3fWftea1VrLQAAwE/tMe4CAABgvhGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjLAHKqqV1XV+nHXAcDsCMkA81RVnVVVXxvj8gV8YMESkgEAoCMkA4xAVa2rqnOr6k+r6i+r6veratG0z3+7qh6squ9X1VumtU9U1WVV9VBVfa+qPlBVe1TVUUkuTHJCVT1eVY8+2/gzqO/tVXVnVT021PjyaXW/u6q+XVWTVfWZqlpUVc9J8gdJXjgs//GqeuGotxvAfCUkA4zOmUlOTvLzSV6S5AND+99OMpHkkCRvS/Kxqnr+8NkFw2cvTvIPkvx6kre01u5McnaSr7fWnttaO+DZxn+2oqrqTUlWD+M+L8kbkjw8bZR/lOSUJEuTvDTJWa21DUlOTfIXw/Kf21r7i9luEIBdlZAMMDq/11q7v7X2SJIPJzljaN+Y5IOttY2ttWuTPJ7kiKraM8npSc5trT3WWluX5N8m+Sdbmvlsx5/mN5L8bmvt5jZlbWvte9M+P7+19hdD3dck+YXtWHeA3YqQDDA6908b/l6Sp7onPNxa2zTts79O8twkByfZexh3+nSHbGX+sx3/KYcmuedZPv/BFmoDWNCEZIDROXTa8GFJttU94UeZOsv8om66B4bhNsvxt+b+THUBma1++QALhpAMMDrvrKrFVXVgkvcn+cyzjdxa+5skn03y4arav6pelOS3kvznYZQfJllcVfvMcPyt+Y9J3l1Vf6emLBum3ZYfJjmoqiZmMC7AbkVIBhidTya5Lsm9mere8K9mMM1vJtkwTPO1YR6XDJ/99yTfSfKDqvrRDMbfotba5zLVR/qTSR5LcmWSA7dVWGvtu0k+leTeqnrU3S2AhaRa89c0gB1VVeuS/EZr7fpx1wLAjnMmGQAAOkIywG6gqi6c9tCP6a8Lx10bwK5IdwsAAOg4kwwAAB0hGQAAOnuNu4AkOfjgg9uSJUvGXQYAALu5W2+99UettZ/d1njzIiQvWbIkt9xyy7jLAABgN1dV35vJeLpbAABAR0gGAICOkAwAAJ150ScZAIDts3Hjxqxfvz5PPPHEuEuZVxYtWpTFixdn77333q7phWQAgF3Y+vXrs//++2fJkiWpqnGXMy+01vLwww9n/fr1Wbp06XbNQ3cLAIBd2BNPPJGDDjpIQJ6mqnLQQQft0Nl1IRkAYBcnID/Tjm4TIRkAgLG58cYb8/rXv37cZTyDPskAALuRJed8aaTzW/fR1410frsKZ5IBANgh69aty5FHHpmzzjorL3nJS3LmmWfm+uuvz4knnpjDDz88N910U2666aaccMIJOe644/KLv/iLueuuu54xnw0bNuStb31rjj/++Bx33HG56qqrxrA2U4RkAAB22Nq1a/Pbv/3b+e53v5vvfve7+eQnP5mvfe1rOe+88/KRj3wkRx55ZP7wD/8wt912Wz74wQ/mfe973zPm8eEPfzgnnXRSbrrpptxwww15z3vekw0bNoxhbcbc3aKqViRZsWzZsnGWMTurJ2Y5/uTc1AEAMI8sXbo0xx57bJLkmGOOyWte85pUVY499tisW7cuk5OTefOb35y77747VZWNGzc+Yx7XXXddrr766px33nlJpu7ccd999+Woo47aqeuSjDkkt9auSXLN8uXL3z7OOgAA2DH77rvv08N77LHH0+/32GOPbNq0KatWrcqrX/3qfPGLX8y6devyqle96hnzaK3liiuuyBFHHLGzyt4q3S0AAJhzk5OTOeSQQ5Ikl1566RbHOfnkk3PBBRektZYkue2223ZWec8gJAMAMOfe+9735txzz81xxx2XTZs2bXGcVatWZePGjXnpS1+aY445JqtWrdrJVf5UPZXUx2n58uXtlltuGXcZM6NPMgAwj9x5551j6bO7K9jStqmqW1try7c1rTPJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAO+T888/PUUcdlTPPPHNO5r969eqnH1W9s4z1sdQAAIzYbJ/psM35bfuZDx//+Mdz/fXXZ/HixaNd9hgt6JC85JwvzXqadYvmoBAAgF3U2WefnXvvvTennnpqTj/99Nxzzz254447snHjxqxevTqnnXZaLr300lx55ZXZsGFD7r777rz73e/Ok08+mU984hPZd999c+211+bAAw/MRRddlDVr1uTJJ5/MsmXL8olPfCL77bffZsu755578s53vjMPPfRQ9ttvv1x00UU58sgjR75eulsAALDdLrzwwrzwhS/MDTfckA0bNuSkk07KTTfdlBtuuCHvec97smHDhiTJHXfckS984Qu5+eab8/73vz/77bdfbrvttpxwwgm57LLLkiRvfOMbc/PNN+f222/PUUcdlYsvvvgZy1u5cmUuuOCC3HrrrTnvvPPyjne8Y07Wa0GfSQYAYHSuu+66XH311U/3H37iiSdy3333JUle/epXZ//998/++++fiYmJrFixIkly7LHH5tvf/naSqSD9gQ98II8++mgef/zxnHzyyZvN//HHH88f/dEf5U1vetPTbT/+8Y/nZF2EZAAARqK1liuuuCJHHHHEZu3f/OY3s++++z79fo899nj6/R577JFNmzYlSc4666xceeWVednLXpZLL700N95442bz+clPfpIDDjgg3/rWt+Z2RaK7BQAAI3LyySfnggsuSGstSXLbbbfNavrHHnssL3jBC7Jx48Zcfvnlz/j8ec97XpYuXZrPfe5zSaZC+e23377jhW+BkAwAwEisWrUqGzduzEtf+tIcc8wxWbVq1aym/9CHPpRXvOIVOfHEE7d6Md7ll1+eiy++OC972ctyzDHH5KqrrhpF6c9QTyX9cVq+fHm75ZZbdvpyt+/uFr82uwlmcNsUAIDtdeedd+aoo44adxnz0pa2TVXd2lpbvq1pnUkGAICOkAwAAB0hGQAAOkIyAMAubj5cYzbf7Og2EZIBAHZhixYtysMPPywoT9Nay8MPP5xFixZt9zw8TAQAYBe2ePHirF+/Pg899NC4S5lXFi1alMWLF2/39EIyAMAubO+9987SpUvHXcZuR3cLAADoCMkAANCZk+4WVfWrSV6X5HlJLm6tXTcXywEAgLkw4zPJVXVJVT1YVXd07adU1V1VtbaqzkmS1tqVrbW3Jzk7yT8ebckAADC3ZtPd4tIkp0xvqKo9k3wsyalJjk5yRlUdPW2UDwyfAwDALmPGIbm19tUkj3TNxydZ21q7t7X2ZJJPJzmtpvxOkj9orf3x6MoFAIC5t6MX7h2S5P5p79cPbb+Z5LVJ/mFVnb2lCatqZVXdUlW3uK8fAADzyZxcuNdaOz/J+dsYZ02SNUmyfPlyj4gBAGDe2NEzyQ8kOXTa+8VDGwAA7LJ2NCTfnOTwqlpaVfskOT3J1TteFgAAjM9sbgH3qSRfT3JEVa2vqre11jYleVeSLye5M8lnW2vfmZtSAQBg55hxn+TW2hlbab82ybXbs/CqWpFkxbJly7ZncgAAmBNjfSx1a+2a1trKiYmJcZYBAACbGWtIBgCA+UhIBgCAjpAMAAAdIRkAADpjDclVtaKq1kxOTo6zDAAA2Iy7WwAAQEd3CwAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjlvAAQBAxy3gAACgo7sFAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEDHLeAAAKDjFnAAANDR3QIAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAICOh4kAAEDHw0QAAKCjuwUAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdDyWGgAAOh5LDQAAHd0tAACgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAACdsYbkqlpRVWsmJyfHWQYAAGxmrCG5tXZNa23lxMTEOMsAAIDN6G4BAAAdIRkAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6Yw3JVbWiqtZMTk6OswwAANjMWENya+2a1trKiYmJcZYBAACb0d0CAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQGXlIrqoXV9XFVfX5Uc8bAAB2hhmF5Kq6pKoerKo7uvZTququqlpbVeckSWvt3tba2+aiWAAA2Blmeib50iSnTG+oqj2TfCzJqUmOTnJGVR090uoAAGAMZhSSW2tfTfJI13x8krXDmeMnk3w6yWkjrg8AAHa6HemTfEiS+6e9X5/kkKo6qKouTHJcVZ27tYmramVV3VJVtzz00EM7UAYAAIzWXqOeYWvt4SRnz2C8NUnWJMny5cvbqOsAAIDttSNnkh9Icui094uHNgAA2KXtSEi+OcnhVbW0qvZJcnqSq0dTFgAAjM9MbwH3qSRfT3JEVa2vqre11jYleVeSLye5M8lnW2vfmbtSAQBg55hRn+TW2hlbab82ybXbu/CqWpFkxbJly7Z3FgAAMHJjfSx1a+2a1trKiYmJcZYBAACbGWtIBgCA+UhIBgCAjpAMAAAdIRkAADpjDclVtaKq1kxOTo6zDAAA2Iy7WwAAQEd3CwAA6AjJAADQEZIBAKAjJAMAQMfdLQAAoOPuFgAA0NHdAgAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEDHfZIBAKDjPskAANDR3QIAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAZ69xLryqViRZsWzZsnGWMRZLzvnSrKdZ99HXzUElAAD03CcZAAA6ulsAAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQMdjqXclq2f5ZMLVk1tsnu0jsT0OGwBYaDyWGgAAOrpbAABAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6Yw3JVbWiqtZMTk6OswwAANjMWENya+2a1trKiYmJcZYBAACb0d0CAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB09hrnwqtqRZIVy5YtG2cZbMvqie2YZnKLzUvO+dKsZrPuo6+b/bJHZbbrvZV1BgB2PWM9k9xau6a1tnJiYjtCGAAAzBHdLQAAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAICOkAwAAJ29Rj3DqnpOko8neTLJja21y0e9DAAAmEszOpNcVZdU1YNVdUfXfkpV3VVVa6vqnKH5jUk+31p7e5I3jLheAACYczPtbnFpklOmN1TVnkk+luTUJEcnOaOqjk6yOMn9w2h/M5oyAQBg55lRSG6tfTXJI13z8UnWttbuba09meTTSU5Lsj5TQXnG8wcAgPlkR/okH5KfnjFOpsLxK5Kcn+T3qup1Sa7Z2sRVtTLJyiQ57LDDdqAMdmurJ2Y5/uQWm5ec86VZL3rdollPMpJlr/vo60azYICFbEz/fizY3+Ej2t7zycgv3GutbUjylhmMtybJmiRZvnx5G3UdAACwvXakO8QDSQ6d9n7x0AYAALu0HQnJNyc5vKqWVtU+SU5PcvVoygIAgPGZ6S3gPpXk60mOqKr1VfW21tqmJO9K8uUkdyb5bGvtO3NXKgAA7Bwz6pPcWjtjK+3XJrl2exdeVSuSrFi2bNn2zgIAAEZurLdoa61d01pbOTExyysiAQBgDrmPMQAAdIRkAADoCMkAANARkgEAoDPWkFxVK6pqzeTk/H80IQAAC4e7WwAAQKdaa+OuIVX1UJLv7aTFHZzkRztpWQuB7Tl6tulo2Z6jZ5uOlu05erbpaO1u2/NFrbWf3dZI8yIk70xVdUtrbfm469hd2J6jZ5uOlu05erbpaNmeo2ebjtZC3Z4u3AMAgI6QDAAAnYUYkteMu4DdjO05erbpaNmeo2ebjpbtOXq26WgtyO254PokAwDAtizEM8kAAPCsFlRIrqpTququqlpbVeeMu55dQVUdWlU3VNWfVtV3quqfDu2rq+qBqvrW8PqVadOcO2zju6rq5PFVPz9V1bqq+pNhu90ytB1YVV+pqruHn88f2quqzh+257er6uXjrX7+qaojpn0Pv1VVf1VV/8x3dOaq6pKqerCq7pjWNuvvZFW9eRj/7qp68zjWZb7Yyjb9N1X13WG7fbGqDhjal1TV/5r2Xb1w2jR/Z/h9sXbY7jWO9Rm3rWzPWR/jcsCUrWzPz0zbluuq6ltD+8L9frbWFsQryZ5J7kny4iT7JLk9ydHjrmu+v5K8IMnLh+H9k/xZkqOTrE7y7i2Mf/SwbfdNsnTY5nuOez3m0yvJuiQHd22/m+ScYficJL8zDP9Kkj9IUklemeSb465/Pr+G4/wHSV7kOzqr7fZLSV6e5I5pbbP6TiY5MMm9w8/nD8PPH/e6zbNt+stJ9hqGf2faNl0yfbxuPjcN27mG7X7quNdtHm3PWR3jcsCzb8/u83+b5F8Mwwv2+7mQziQfn2Rta+3e1tqTST6d5LQx1zTvtda+31r742H4sSR3JjnkWSY5LcmnW2s/bq39eZK1mdr2PLvTkvynYfg/JfnVae2XtSnfSHJAVb1gHAXuIl6T5J7W2rM9nMh3tNNa+2qSR7rm2X4nT07yldbaI621v0zylSSnzH3189OWtmlr7brW2qbh7TeSLH62eQzb9XmttW+0qURyWX66HxaUrXxHt2Zrx7gcMHi27TmcDf5HST71bPNYCN/PhRSSD0ly/7T36/PsYY9OVS1JclySbw5N7xr+bHjJU3+Kje08Ey3JdVV1a1WtHNp+rrX2/WH4B0l+bhi2PWfn9Gz+i913dPvN9jtpu87OWzN15u0pS6vqtqr6H1X194e2QzK1HZ9imz7TbI5x39GZ+ftJfthau3ta24L8fi6kkMwOqKrnJrkiyT9rrf1Vkv+Q5OeT/EKS72fqTzPMzN9rrb08yalJ3llVvzT9w+F/5G47M0tVtU+SNyT53NDkOzoivpOjVVXvT7IpyeVD0/eTHNZaOy7JbyX5ZFU9b1z17UIc43PjjGx+smHBfj8XUkh+IMmh094vHtrYhqraO1MB+fLW2heSpLX2w9ba37TWfpLkovz0z9W28za01h4Yfj6Y5IuZ2nY/fKobxfDzwWF023PmTk3yx621Hya+oyMw2++k7ToDVXVWktcnOXP4z0eGbgEPD8O3Zqrf7Esytf2md8mwTafZjmPcd3QbqmqvJG9M8pmn2hby93MhheSbkxxeVUuHM06nJ7l6zDXNe0PfpIuT3Nla+3fT2qf3i/3fkzx1hezVSU6vqn2rammSwzPVsZ8kVfWcqtr/qeFMXchzR6a221N3A3hzkquG4auT/PpwR4FXJpmc9idwNrfZ2Q/f0R022+/kl5P8clU9f/iz9y8PbQyq6pQk703yhtbaX09r/9mq2nMYfnGmvpP3Dtv1r6rqlcPv4l/PT/fDgrcdx7gcsG2vTfLd1trT3SgW8vdzr3EXsLO01jZV1bsy9Ut7zySXtNa+M+aydgUnJvknSf7kqdvBJHlfkjOq6hcy9SfYdUn+jyRprX2nqj6b5E8z9efEd7bW/manVz1//VySLw53ydkrySdba/+1qm5O8tmqeluS72XqookkuTZTdxNYm+Svk7xl55c8/w3/4fjfMnwPB7/rOzozVfWpJK9KcnBVrU/yL5N8NLP4TrbWHqmqD2UqiCTJB1trM73QarezlW16bqbuuPCV4XfAN1prZ2fqTgMfrKqNSX6S5Oxp2+4dSS5N8jOZ6sM8vR/zgrGV7fmq2R7jcsCULW3P1trFeeZ1HckC/n564h4AAHQWUncLAACYESEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZYIaq6lVVtX7MNZxVVV+b9v7xqnrxMHxpVf2r8VUHsPsQkgF2gj7cjkpr7bmttXtHPd8dUVVLqqpV1V7jrgVgewnJAADQEZIBOlW1rqrOrao/raq/rKrfr6pF0z7/7ap6sKq+X1VvmdY+UVWXVdVDVfW9qvpAVe1RVUcluTDJCUP3iEefbfxZ1tqqatkW2vevqhuq6vyasm9VnVdV91XVD6vqwqr6mRnM/7Sq+lZV/VVV3VNVpwztN1bVh6rqf1bVY1V1XVUdPEz21eHno8P6njCbdQKYD4RkgC07M8nJSX4+yUuSfGBo/9tJJpIckuRtST5WVc8fPrtg+OzFSf5Bkl9P8pbW2p1Jzk7y9aF7xAHPNv6OFl5VByX5b0n+Z2vt/2qttSQfHdbjF5IsG+r/F9uYz/FJLkvyniQHJPmlJOumjfJrQ71/K8k+Sd49tP/S8POAYX2/vqPrBLCzCckAW/Z7rbX7W2uPJPlwkjOG9o1JPtha29hauzbJ40mOqKo9k5ye5NzW2mOttXVJ/m2Sf7Klmc92/Fl4YZL/keRzrbUPDMuqJCuT/PPW2iOttceSfGRY/rN5W5JLWmtfaa39pLX2QGvtu9M+//3W2p+11v5Xks9mKoAD7BZcVAGwZfdPG/5epsJnkjzcWts07bO/TvLcJAcn2XsYd/p0h2xl/rMdf6Zel6ngfuG0tp9Nsl+SW6fycpKkkuy5jXkdmuTaZ/n8B9OGn9oOALsFZ5IBtuzQacOHJfmLbYz/o0ydZX5RN90Dw3Cb5fjb66Ik/zXJtVX1nGnL+l9JjmmtHTC8Jlpr2wq192equ8ls9esKsMsRkgG27J1VtbiqDkzy/iSfebaRW2t/k6kuBx8eLpp7UZLfSvKfh1F+mGRxVe0zw/F3xLuS3JXkmqr6mdbaTzIVnv99Vf2tJKmqQ6rq5G3M5+Ikb6mq1wwXIB5SVUfOYPkPJflJpvpaA+yShGSALftkkuuS3JvkniQzeUjHbybZMEzztWEelwyf/fck30nyg6r60QzG327DhXork6xPctVwZ47/O8naJN+oqr9Kcn2SI7Yxn5sydWHev08ymam+zi96tmmG6f46U/24/2dVPVpVr9yB1QEYi5r6XQrAU6pqXZLfaK1dP+5aABgPZ5IBAKAjJAPMM8ODPh7fwuvCbU8962W9byvL+oNRLwtgV6K7BQAAdJxJBgCAjpAMAACdefHEvYMPPrgtWbJk3GUAALCbu/XWW3/UWvvZbY03L0LykiVLcsstt4y7DAAAdnNV9b2ZjKe7BQAAdIRkAADoCMkAANCZF32SAQDYPhs3bsz69evzxBNPjLuUeWXRokVZvHhx9t577+2aXkgGANiFrV+/Pvvvv3+WLFmSqhp3OfNCay0PP/xw1q9fn6VLl27XPHS3AADYhT3xxBM56KCDBORpqioHHXTQDp1dF5IBAHZxAvIz7eg2EZIBABibG2+8Ma9//evHXcYz6JMMALAbWXLOl0Y6v3Uffd1I57ercCYZAIAdsm7duhx55JE566yz8pKXvCRnnnlmrr/++px44ok5/PDDc9NNN+Wmm27KCSeckOOOOy6/+Iu/mLvuuusZ89mwYUPe+ta35vjjj89xxx2Xq666agxrM2VBn0nenv9pLdT/TQEAPJu1a9fmc5/7XC655JL83b/7d/PJT34yX/va13L11VfnIx/5SC677LL84R/+Yfbaa69cf/31ed/73pcrrrhis3l8+MMfzkknnZRLLrkkjz76aI4//vi89rWvzXOe85ydvj4LOiQDADAaS5cuzbHHHpskOeaYY/Ka17wmVZVjjz0269aty+TkZN785jfn7rvvTlVl48aNz5jHddddl6uvvjrnnXdekqk7d9x333056qijduq6JEIyAAAjsO+++z49vMceezz9fo899simTZuyatWqvPrVr84Xv/jFrFu3Lq961aueMY/WWq644oocccQRO6vsrdInGQCAOTc5OZlDDjkkSXLppZducZyTTz45F1xwQVprSZLbbrttZ5X3DEIyAABz7r3vfW/OPffcHHfccdm0adMWx1m1alU2btyYl770pTnmmGOyatWqnVzlT9VTSX2cli9f3m655ZadvtztunBv0a/NboLVk7NeBgDATN15551j6bO7K9jStqmqW1try7c1rTPJAADQEZIBAKAjJAMAQGdObgFXVb+a5HVJnpfk4tbadXOxHAAAmAszPpNcVZdU1YNVdUfXfkpV3VVVa6vqnCRprV3ZWnt7krOT/OPRlgwAAHNrNt0tLk1yyvSGqtozyceSnJrk6CRnVNXR00b5wPA5AACCrJ4LAAAgAElEQVTsMmYckltrX03ySNd8fJK1rbV7W2tPJvl0ktNqyu8k+YPW2h+PrlwAAOab888/P0cddVTOPPPMOZn/6tWrn35U9c6yo32SD0ly/7T365O8IslvJnltkomqWtZau7CfsKpWJlmZJIcddtgOlgEAQJJk9cSI57ftZz58/OMfz/XXX5/FixePdtljNCcX7rXWzk9y/jbGWZNkTTL1MJG5qAMAgLl19tln5957782pp56a008/Pffcc0/uuOOObNy4MatXr85pp52WSy+9NFdeeWU2bNiQu+++O+9+97vz5JNP5hOf+ET23XffXHvttTnwwANz0UUXZc2aNXnyySezbNmyfOITn8h+++232fLuueeevPOd78xDDz2U/fbbLxdddFGOPPLIka/Xjt4C7oEkh057v3hoAwBgAbjwwgvzwhe+MDfccEM2bNiQk046KTfddFNuuOGGvOc978mGDRuSJHfccUe+8IUv5Oabb8773//+7Lfffrnttttywgkn5LLLLkuSvPGNb8zNN9+c22+/PUcddVQuvvjiZyxv5cqVueCCC3LrrbfmvPPOyzve8Y45Wa8dPZN8c5LDq2pppsLx6Ulm+dxmAAB2B9ddd12uvvrqp/sPP/HEE7nvvvuSJK9+9auz//77Z//998/ExERWrFiRJDn22GPz7W9/O8lUkP7ABz6QRx99NI8//nhOPvnkzeb/+OOP54/+6I/ypje96em2H//4x3OyLjMOyVX1qSSvSnJwVa1P8i9baxdX1buSfDnJnkkuaa19Z04qBQBgXmut5YorrsgRRxyxWfs3v/nN7Lvvvk+/32OPPZ5+v8cee2TTpk1JkrPOOitXXnllXvayl+XSSy/NjTfeuNl8fvKTn+SAAw7It771rbldkczu7hZntNZe0Frbu7W2uLV28dB+bWvtJa21n2+tfXg2C6+qFVW1ZnJy2x3CAQCY304++eRccMEFaW3qcrPbbrttVtM/9thjecELXpCNGzfm8ssvf8bnz3ve87J06dJ87nOfSzIVym+//fYdL3wLxvpY6tbaNa21lRMTI74KEwCAnW7VqlXZuHFjXvrSl+aYY47JqlWrZjX9hz70obziFa/IiSeeuNWL8S6//PJcfPHFednLXpZjjjkmV1111ShKf4Z6KumP0/Lly9stt9yy05e75JwvzXqadYtm2eV6BrdNAQDYXnfeeWeOOuqocZcxL21p21TVra215duadqxnkgEAYD4SkgEAoCMkAwBAR0gGANjFzYdrzOabHd0mYw3JbgEHALBjFi1alIcfflhQnqa1locffjiLFi3a7nns6BP3dkhr7Zok1yxfvvzt46wDAGBXtXjx4qxfvz4PPfTQuEuZVxYtWpTFixdv9/RjDckAAOyYvffeO0uXLh13GbsdfZIBAKAjJAMAQEdIBgCAjpAMAAAdt4ADAIDOWENya+2a1trKiYmJcZYBAACb0d0CAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQMd9kgEAoOM+yQAA0NHdAgAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHU/cAwCAjifuAQBAR3cLAADoCMkAANARkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BlrSK6qFVW1ZnJycpxlAADAZsYakltr17TWVk5MTIyzDAAA2IzuFgAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAACdvca58KpakWTFsmXLxlnGrmP1xCzHn5ybOgAAdnNjPZPcWrumtbZyYmKW4Q8AAOaQ7hYAANARkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAzlhDclWtqKo1k5OT4ywDAAA2M9aQ3Fq7prW2cmJiYpxlAADAZnS3AACAjpAMAAAdIRkAADp7jbsAdgGrt6PP+GoXYwIAuy5nkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADp7jbuAhWrJOV+a9TTrFs1BIQAAPIMzyQAA0BGSAQCgIyQDAEBHn+QFaLb9ofWFBgAWmpGfSa6qF1fVxVX1+VHPGwAAdoYZheSquqSqHqyqO7r2U6rqrqpaW1XnJElr7d7W2tvmolgAANgZZnom+dIkp0xvqKo9k3wsyalJjk5yRlUdPdLqAABgDGYUkltrX03ySNd8fJK1w5njJ5N8OslpI64PAAB2uh25cO+QJPdPe78+ySuq6qAkH05yXFWd21r711uauKpWJlmZJIcddtgOlAFzZPXELMefnJs6AICdbuR3t2itPZzk7BmMtybJmiRZvnx5G3UdAACwvXbk7hYPJDl02vvFQxsAAOzSdiQk35zk8KpaWlX7JDk9ydWjKQsAAMZnpreA+1SSryc5oqrWV9XbWmubkrwryZeT3Jnks62178xdqQAAsHPMqE9ya+2MrbRfm+TakVYEAABjNvIn7s1GVa2oqjWTk+4KAADA/DHWkNxau6a1tnJiYpa32gIAgDk01pAMAADzkZAMAAAdIRkAADpCMgAAdEb+WOrZqKoVSVYsW7ZsnGUwn62e5UWdq3eDO6XMdp2T3WO9AWAecXcLAADo6G4BAAAdIRkAADpCMgAAdIRkAADoCMkAANAZa0iuqhVVtWZy0u2rAACYP9wCDgAAOrpbAABAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0NlrnAuvqhVJVixbtmycZQBPWb0d9yxf7WFAAOx+PEwEAAA6ulsAAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQ8TARdqol53xpVuOvWzSe5Y5y2QDArsfDRAAAoKO7BQAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOiMNSRX1YqqWjM5OTnOMgAAYDNjDcmttWtaaysnJibGWQYAAGxGdwsAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAACdvca58KpakWTFsmXLxlkGMB+snpjl+JO7x7IBmJfGeia5tXZNa23lxMQs/4ECAIA5pLsFAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANDZa5wLr6oVSVYsW7ZsnGUAc2DJOV+a1fjrFo1nuaNcNjO0emKW40/OTR0Az2KsZ5Jba9e01lZOTMzyFyYAAMwh3S0AAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADo7DXqGVbVc5J8PMmTSW5srV0+6mUAAMBcmtGZ5Kq6pKoerKo7uvZTququqlpbVecMzW9M8vnW2tuTvGHE9QIAwJybaXeLS5OcMr2hqvZM8rEkpyY5OskZVXV0ksVJ7h9G+5vRlAkAADvPjEJya+2rSR7pmo9Psra1dm9r7ckkn05yWpL1mQrKM54/AADMJzvSJ/mQ/PSMcTIVjl+R5Pwkv1dVr0tyzdYmrqqVSVYmyWGHHbYDZQBbs+ScL81q/HWL5qiQBWL22/vXZr+Q1ZOzn2aL85kYz3LHbTdY7zn/ns3Ddd6lzPY7lszLbe57NgcX7rXWNiR5ywzGW5NkTZIsX768jboOAADYXjvSHeKBJIdOe794aAMAgF3ajoTkm5McXlVLq2qfJKcnuXo0ZQEAwPjM9BZwn0ry9SRHVNX6qnpba21Tkncl+XKSO5N8trX2nbkrFQAAdo4Z9UlurZ2xlfZrk1w70ooAAGDMxnqLtqpaUVVrJifn/xWOAAAsHGMNya21a1prKycmtuN2KQAAMEc87AMAADpCMgAAdIRkAADoCMkAANBxdwsAAOhUa23cNaSqHkryvZ20uIOT/GgnLYvZsW/mL/tmfrJf5i/7Zv6yb+annblfXtRa+9ltjTQvQvLOVFW3tNaWj7sOnsm+mb/sm/nJfpm/7Jv5y76Zn+bjftEnGQAAOkIyAAB0FmJIXjPuAtgq+2b+sm/mJ/tl/rJv5i/7Zn6ad/tlwfVJBgCAbVmIZ5IBAOBZLaiQXFWnVNVdVbW2qs4Zdz27u6o6tKpuqKo/rarvVNU/HdoPrKqvVNXdw8/nD+1VVecP++fbVfXyafN68zD+3VX15nGt0+6mqvasqtuq6r8M75dW1TeHffCZqtpnaN93eL92+HzJtHmcO7TfVVUnj2dNdh9VdUBVfb6qvltVd1bVCY6Z+aGq/vnwu+yOqvpUVS1yzIxHVV1SVQ9W1R3T2kZ2nFTV36mqPxmmOb+qaueu4a5rK/vm3wy/075dVV+sqgOmfbbF42FrmW1rx9ycaK0tiFeSPZPck+TFSfZJcnuSo8dd1+78SvKCJC8fhvdP8mdJjk7yu0nOGdrPSfI7w/CvJPmDJJXklUm+ObQfmOTe4efzh+Hnj3v9dodXkt9K8skk/2V4/9kkpw/DFyb5P4fhdyS5cBg+PclnhuGjh2Np3yRLh2Nsz3Gv1678SvKfkvzGMLxPkgMcM+N/JTkkyZ8n+Znh/WeTnOWYGdv++KUkL09yx7S2kR0nSW4axq1h2lPHvc67ymsr++aXk+w1DP/OtH2zxeMhz5LZtnbMzcVrIZ1JPj7J2tbava21J5N8OslpY65pt9Za+35r7Y+H4ceS3Jmpf2hOy1QQyPDzV4fh05Jc1qZ8I8kBVfWCJCcn+Upr7ZHW2l8m+UqSU3biquyWqmpxktcl+Y/D+0pyUpLPD6P0++apffb5JK8Zxj8tyadbaz9urf15krWZOtbYDlU1kal/YC5Oktbak621R+OYmS/2SvIzVbVXkv2SfD+OmbForX01ySNd80iOk+Gz57XWvtGmkthl0+bFNmxp37TWrmutbRrefiPJ4mF4a8fDFjPbNv6dGrmFFJIPSXL/tPfrhzZ2guFPjccl+WaSn2utfX/46AdJfm4Y3to+su/mxv+b5L1JfjK8PyjJo9N+kU3fzk/vg+HzyWF8+2a0liZ5KMnvD91g/mNVPSeOmbFrrT2Q5Lwk92UqHE8muTWOmflkVMfJIcNw385ovDVTZ+eT2e+bZ/t3auQWUkhmTKrquUmuSPLPWmt/Nf2z4X/pbrGyk1XV65M82Fq7ddy1sJm9MvVnyv/QWjsuyYZM/dn4aY6Z8Rj6t56Wqf/IvDDJc+Ls/LzlOJmfqur9STYluXzctczEQgrJDyQ5dNr7xUMbc6iq9s5UQL68tfaFofmHw5+zMvx8cGjf2j6y70bvxCRvqKp1mfoz1klJ/r9M/Rlyr2Gc6dv56X0wfD6R5OHYN6O2Psn61to3h/efz1RodsyM32uT/Hlr7aHW2sYkX8jUceSYmT9GdZw8kJ92B5jezg6oqrOSvD7JmcN/YpLZ75uHs/VjbuQWUki+Ocnhw1WR+2TqQoqrx1zTbm3oO3Rxkjtba/9u2kdXJ3nqKuI3J7lqWvuvD1civzLJ5PCnsy8n+eWqev5wNueXhza2U2vt3Nba4tbakkwdC/+9tXZmkhuS/MNhtH7fPLXP/uEwfhvaTx+u5F+a5PBMXfDCdmit/SDJ/VV1xND0miR/GsfMfHBfkldW1X7D77an9o1jZv4YyXEyfPZXVfXKYV//+rR5sR2q6pRMde97Q2vtr6d9tLXjYYuZbTiGtnbMjd5cXRE4H1+ZusL1zzJ1xeT7x13P7v5K8vcy9eeubyf51vD6lUz1KfpvSe5Ocn2SA4fxK8nHhv3zJ0mWT5vXWzPVoX9tkreMe912p1eSV+Wnd7d4caZ+Qa1N8rkk+w7ti4b3a4fPXzxt+vcP++yuuAJ8FPvjF5LcMhw3V2bqqnvHzDx4Jfl/knw3yR1JPpGpK/IdM+PZF5/KVN/wjZn6C8zbRnmcJFk+7Od7kvxehoeveW33vvn/27v/aL2qOj/8708AiY54UXAcJWLiBCFkQUyNKDKuAo4LGMzA8ke/OkxH1E6Wy3E6ba02oGkZLdauRbXyw2HFgTK0qIOiQganw6Chav0RYKJWBSqwIsaqaNA7EAUT2d8/7gNedhK4N7n3Pjfc12utu/KcffY553MeNuTNufucc3vG5hg/lAUuHtd/p/8+ZBeZbVf/zk3HjzfuAQBAZy5NtwAAgAkRkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANARkoE5qaqOr6rNQ67hzKr64rjl+6rquYPPl1XVfxxedZNTVZuq6ncHn8+uqr8cdk0Ae2LfYRcAsLepqjOT/IvW2u9M5X5ba0+eyv0NS2vtvcOuAWBPuZIMAAAdIRl4XBtMAzirqr5dVT+tqv9WVfPHrX9bVd1dVT+oqjeMax+pqsur6sdV9d2qeldVzauqJUkuTnLsYHrEzx6t/yRrbVW1eCftB1TV+qo6v8bsX1XnVdVdVfWjqrq4qp74GPs+vqo2V9U7xp3v6VX1e1X1f6vqnqo6e1z/eVW1uqruqKotVXVlVT1t3Pp/PjjPLVX1zu5Y51TV/xi3/PGq+mFVjVbV56tq6bh1l1XVRVV1bVXdW1Vfrarfnsz3BjAdhGRgLjgjyUlJfjvJ85K8a9D+W0lGkhyS5E1JLqqqpw7WXTBY99wk/zTJHyV5Q2vtliRvTvLl1tqTW2sHPlr/PS28qg5K8tkk/7u19i9bay3J+wbn8fwkiwf1//sJ7O63kswf1//DSf4wyQuSvDTJmqpaNOj7p0lOH5zLs5L8NMlFg5qOTPIXSf75YN1BSRY8ynH/NslhSX4zyT8kuaJb/9okf57kqUluT3LuBM4FYFoJycBccGFr7XuttXsyFsBeN2jfluTdrbVtrbXPJLkvyeFVtU/GgttZrbV7W2ubkvyXjIXCHUy2/yQ8K8n/SvLx1tq7BseqJKuS/OvW2j2ttXuTvHdw/MeyLcm5rbVtST6W5OAkHxzU/K0k306ybND3zUne2Vrb3Fp7IMk5SV5dVfsmeXWSv2mtfX6wbk2SB3d10NbapYNjPLSfZVU1Mq7Lp1prG1pr2zMWoJ8/gXMBmFZu3APmgu+N+/zdjIXPJNkyCGYP+XmSJ2csPO436Dt+u0N2sf/J9p+oUzMW3C8e1/b0JE9KcvNYXk6SVJJ9JrC/La21Xw0+/2Lw54/Grf9Fxs4/SZ6T5FNVNT78/irJMzL2/T38nbbWtlbVlp0dcPA/EOcmec2g9of2d3CS0cHnH47b5KF/BgBD5UoyMBc8e9znQ5P8v8fo/5OMXXV9Trfd9wef2yT7764PJ/mfST5TVb8x7li/SLK0tXbg4GdkGp6M8b0kp4w7xoGttfmtte8n+UHGfadV9aSMTbnYmT9IclqS383YdJSFD202xfUCTCkhGZgL/qSqFgxuPHtnkr9+tM6Dq61XJjl3cNPcc5L8myQP3Yz2oyQLquoJE+y/J96a5LYk66rqia21BzMWnj9QVb+ZJFV1SFWdNAXHGu/ijJ3PcwbHeHpVnTZY94kkr6iq3xl8B+/Orv8+OSDJA0m2ZOwKuMfDAXsFIRmYCz6S5Lokdya5I8lEXtLxp0m2Drb54mAflw7WfS7Jt5L8sKp+MoH+u21wo96qJJuTXD14Mse/y9gNbl+pqn9Mcn2Sw/f0WJ0PJrkmyXVVdW+SryR50aCmbyX5k4yd4w8ydlPfrl7McnnGpp58P2Nznr8yxXUCTIsa++8vwONTVW3K2Is/rh92LQDsPVxJBgCAjpAMMI0GL/q4byc/Fz/21pM+1tm7ONbfTvWxAB7vTLcAAICOK8kAANARkgEAoDMr3rh38MEHt4ULFw67DAAAHuduvvnmn7TWnv5Y/WZFSF64cGFuuummYZcBAMDjXFV9dyL9TLcAAICOkAwAAJ2hhuSqWllVa0dHR4dZBgAAPMJQ5yS31tYlWbdixYo/HmYdAAB7q23btmXz5s25//77h13KrDJ//vwsWLAg++23325tPytu3AMAYPds3rw5BxxwQBYuXJiqGnY5s0JrLVu2bMnmzZuzaNGi3dqHOckAAHux+++/PwcddJCAPE5V5aCDDtqjq+tCMgDAXk5A3tGefidCMgAAQ3PDDTfkFa94xbDL2IE5yQAAjyMLV187pfvb9L5Tp3R/ewuPgAMAYI9s2rQpRxxxRM4888w873nPyxlnnJHrr78+xx13XA477LBs2LAhGzZsyLHHHpvly5fnJS95SW677bYd9rN169a88Y1vzDHHHJPly5fn6quvHsLZjBlqSG6trWutrRoZGRlmGQAA7KHbb789b3vb23Lrrbfm1ltvzUc+8pF88YtfzHnnnZf3vve9OeKII/KFL3whGzduzLvf/e6cffbZO+zj3HPPzYknnpgNGzZk/fr1efvb356tW7cO4Wzm+HSL3fl1xFz9lQMAwKNZtGhRjjrqqCTJ0qVL87KXvSxVlaOOOiqbNm3K6OhoXv/61+c73/lOqirbtm3bYR/XXXddrrnmmpx33nlJxp7ccdddd2XJkiUzei7JHA/JAABMjf333//hz/PmzXt4ed68edm+fXvWrFmTE044IZ/61KeyadOmHH/88Tvso7WWq666KocffvhMlb1Lnm4BAMC0Gx0dzSGHHJIkueyyy3ba56STTsoFF1yQ1lqSZOPGjTNV3g6EZAAApt073vGOnHXWWVm+fHm2b9++0z5r1qzJtm3bcvTRR2fp0qVZs2bNDFf5a/VQUh+mFStWtJtuumnGj2tOMgCwt7vllluGMmd3b7Cz76aqbm6trXisbV1JBgCAjuckAwBAx3OSAQCgY7oFAAB0hGQAAOgIyQAA0BGSAQDYI+eff36WLFmSM844Y1r2f8455zz8quqZ4rXUAACPJ+dM8QMRznnsp5B96EMfyvXXX58FCxZM7bGHyJVkAAB225vf/ObceeedOeWUU3LuuefmjW98Y4455pgsX748V199dZKx11CffvrpefnLX56FCxfmwgsvzPvf//4sX748L37xi3PPPfckST784Q/nhS98YZYtW5ZXvepV+fnPf77D8e64446cfPLJecELXpCXvvSlufXWW6flvIRkAAB228UXX5xnPetZWb9+fbZu3ZoTTzwxGzZsyPr16/P2t789W7duTZJ885vfzCc/+cnceOONeec735knPelJ2bhxY4499thcfvnlSZJXvvKVufHGG/P1r389S5YsySWXXLLD8VatWpULLrggN998c84777y85S1vmZbzGup0i6pamWTl4sWLh1kGAABT4Lrrrss111zz8Pzh+++/P3fddVeS5IQTTsgBBxyQAw44ICMjI1m5cmWS5Kijjso3vvGNJGNB+l3veld+9rOf5b777stJJ530iP3fd999+dKXvpTXvOY1D7c98MAD03IuQw3JrbV1SdatWLHij4dZBwAAe661lquuuiqHH374I9q/+tWvZv/99394ed68eQ8vz5s3L9u3b0+SnHnmmfn0pz+dZcuW5bLLLssNN9zwiP08+OCDOfDAA/O1r31tek8kbtzbu0x2Iv4EJtoDAEyVk046KRdccEEuuOCCVFU2btyY5cuXT3j7e++9N8985jOzbdu2XHHFFTnkkEMesf4pT3lKFi1alI9//ON5zWtek9ZavvGNb2TZsmVTfSrmJAMAMDXWrFmTbdu25eijj87SpUuzZs2aSW3/nve8Jy960Yty3HHH5YgjjthpnyuuuCKXXHJJli1blqVLlz58c+BUq9batOx4MlasWNFuuummGT/uwtXXTnqbTe87dRoqmSBXkgGAzi233JIlS5YMu4xZaWffTVXd3Fpb8VjbupIMAAAdIRkAADpCMgAAdIRkAIC93Gy4x2y22dPvxCPgJsvNcwDALDJ//vxs2bIlBx10UKpq2OXMCq21bNmyJfPnz9/tfQjJAAB7sQULFmTz5s358Y9/POxSZpX58+dnwYIFu72911IDAOzF9ttvvyxatGjYZTzuDHVOcmttXWtt1cjIJKcwAADANHLjHgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAICOl4nw2Cb7lsHEmwYBgL2akDwkC1dfO+ltNu3+mxUBAJgE0y0AAKAjJAMAQEdIBgCAzlBDclWtrKq1o6Nu8gIAYPYYakhura1rra0aGdmNpycAAMA0Md0CAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAIDOUENyVa2sqrWjo6PDLAMAAB5hqCG5tbautbZqZGRkmGUAAMAjmG4BAAAdIRkAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOjcjS5MAACAASURBVEIyAAB0hGQAAOgIyQAA0BGSAQCgs++wC4BHdc7IJPuPTk8dAMCc4koyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAnaGG5KpaWVVrR0c92xYAgNljqCG5tbautbZqZGSSL4wAAIBpZLoFAAB0vJZ6Dlq4+tpJ9d80f5oKme28EhsA5ixXkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgM+UhuaqeW1WXVNUnpnrfAAAwEyYUkqvq0qq6u6q+2bWfXFW3VdXtVbU6SVprd7bW3jQdxQIAwEyY6JXky5KcPL6hqvZJclGSU5IcmeR1VXXklFYHAABDMKGQ3Fr7fJJ7uuZjktw+uHL8yyQfS3LaFNcHAAAzbt892PaQJN8bt7w5yYuq6qAk5yZZXlVntdb+0842rqpVSVYlyaGHHroHZbA3Wbj62kn13zR/mgoBAHgUexKSd6q1tiXJmyfQb22StUmyYsWKNtV1AADA7tqTp1t8P8mzxy0vGLQBAMBebU9C8o1JDquqRVX1hCSvTXLN1JQFAADDM9FHwH00yZeTHF5Vm6vqTa217UnemuTvktyS5MrW2remr1QAAJgZE5qT3Fp73S7aP5PkM1NaEQAADJnXUgMAQGeoIbmqVlbV2tHR0WGWAQAAjzDUkNxaW9daWzUyMjLMMgAA4BFMtwAAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6HgEHAAAdDwCDgAAOqZbAABAR0gGAICOkAwAAB0hGQAAOkIyAAB0PAIOAAA6HgEHAAAd0y0AAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANDxxj0AAOh44x4AAHRMtwAAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQGffYR68qlYmWbl48eJhlsEcsHD1tZPeZtP8aSgEANgreOMeAAB0TLcAAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHT2HebBq2plkpWLFy8eZhkwrRauvnZS/Te979ThHXv+H0z+IOeMTn6bne5nZDjHBYCdGOqV5NbautbaqpGRSf7lCAAA08h0CwAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOvsO8+BVtTLJysWLFw+zDJhdzhnZjW1Gp74OAJjDhnolubW2rrW2amRkN0IBAABME9MtAACgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hhqSq2plVa0dHR0dZhkAAPAIQw3JrbV1rbVVIyMjwywDAAAewXQLAADoCMkAANARkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgs+9U77CqfiPJh5L8MskNrbUrpvoYAAAwnSZ0JbmqLq2qu6vqm137yVV1W1XdXlWrB82vTPKJ1tofJ/n9Ka4XAACm3USnW1yW5OTxDVW1T5KLkpyS5Mgkr6uqI5MsSPK9QbdfTU2ZAAAwcyYUkltrn09yT9d8TJLbW2t3ttZ+meRjSU5LsjljQXnC+wcAgNlkT+YkH5JfXzFOxsLxi5Kcn+TCqjo1ybpdbVxVq5KsSpJDDz10D8oAZqOFq6+dVP9N86epkIk4Z2SS/Ud32jzpc37fqZM7LgzZXjXGp+jf67lqr/pnPU2m/Ma91trWJG+YQL+1SdYmyYoVK9pU1wEAALtrT6ZDfD/Js8ctLxi0AQDAXm1PQvKNSQ6rqkVV9YQkr01yzdSUBQAAwzPRR8B9NMmXkxxeVZur6k2tte1J3prk75LckuTK1tq3pq9UAACYGROak9xae90u2j+T5DNTWhEAAAzZUB/RVlUrq2rt6Kg7SgEAmD2GGpJba+taa6tGRib5mBYAAJhGXvYBAAAdIRkAADpCMgAAdIRkAADoVGvDfyN0Vf04yXdn6HAHJ/nJDB2LxwdjhskyZpgsY4bJMmZ233Naa09/rE5DDclVtTLJyiTrWmvrZuiYN7XWVszEsXh8MGaYLGOGyTJmmCxjZvpN6GUi02UQjGckHAMAwESZkwwAAJ25GJLXDrsA9jrGDJNlzDBZxgyTZcxMs1lx4x4AAMwmc/FKMgAAPKo5FZKr6uSquq2qbq+q1cOuh9mhqi6tqrur6pvj2p5WVX9fVd8Z/PnUQXtV1fmDMfSNqvonw6ucYamqZ1fV+qr6dlV9q6r+bNBu3LBTVTW/qjZU1dcHY+bPB+2Lquqrg7Hx11X1hEH7/oPl2wfrFw6zfoanqvapqo1V9TeDZWNmhsyZkFxV+yS5KMkpSY5M8rqqOnK4VTFLXJbk5K5tdZLPttYOS/LZwXIyNn4OG/ysSvIXM1Qjs8v2JG9rrR2Z5MVJ/mTw3xPjhl15IMmJrbVlSZ6f5OSqenGS/5zkA621xUl+muRNg/5vSvLTQfsHBv2Ym/4syS3jlo2ZGTJnQnKSY5Lc3lq7s7X2yyQfS3LakGtiFmitfT7JPV3zaUn+avD5r5KcPq798jbmK0kOrKpnzkylzBattR+01v5h8PnejP0FdkiMG3Zh8M/+vsHifoOfluTEJJ8YtPdj5qGx9IkkL6uqmqFymSWqakGSU5P85WC5YszMmLkUkg9J8r1xy5sHbbAzz2it/WDw+YdJnjH4bBzxCINfaS5P8tUYNzyKwa/Nv5bk7iR/n+SOJD9rrW0fdBk/Lh4eM4P1o0kOmtmKmQX+a5J3JHlwsHxQjJkZM5dCMuyWNvYIGI+BYQdV9eQkVyX5V621fxy/zrih11r7VWvt+UkWZOy3m0cMuSRmsap6RZK7W2s3D7uWuWouheTvJ3n2uOUFgzbYmR899OvwwZ93D9qNI5IkVbVfxgLyFa21Tw6ajRseU2vtZ0nWJzk2Y1NvHnr77fhx8fCYGawfSbJlhktluI5L8vtVtSljU0RPTPLBGDMzZi6F5BuTHDa4K/QJSV6b5Joh18TsdU2S1w8+vz7J1ePa/2jwtIIXJxkd9+t15ojBPL9LktzSWnv/uFXGDTtVVU+vqgMHn5+Y5OUZm8u+PsmrB936MfPQWHp1ks81LzaYU1prZ7XWFrTWFmYss3yutXZGjJkZM6deJlJVv5ex+T37JLm0tXbukEtiFqiqjyY5PsnBSX6U5D8k+XSSK5McmuS7Sf5Za+2eQTi6MGNPw/h5kje01m4aRt0MT1X9TpIvJPk/+fVcwbMzNi/ZuGEHVXV0xm6q2idjF6iubK29u6qem7GrhE9LsjHJH7bWHqiq+Un+e8bmu9+T5LWttTuHUz3DVlXHJ/m3rbVXGDMzZ06FZAAAmIi5NN0CAAAmREgGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGGKiq46tq85BrOLOqvjhu+b6qeu7g82VV9R+HVx3A3CEkA0yBPtxOldbak1trd071fgF4dEIyAAB0hGRgzqmqTVV1VlV9u6p+WlX/rarmj1v/tqq6u6p+UFVvGNc+UlWXV9WPq+q7VfWuqppXVUuSXJzk2MH0iJ89Wv9J1tqqavFO2g+oqvVVdX6N2b+qzququ6rqR1V1cVU98TH2fXxVba6qd4w739Or6veq6v9W1T1Vdfa4/vOqanVV3VFVW6rqyqp62rj1H6+qH1bVaFV9vqqWjlt3WVVdVFXXVtW9VfXVqvrtyXwXADNJSAbmqjOSnJTkt5M8L8m7Bu2/lWQkySFJ3pTkoqp66mDdBYN1z03yT5P8UZI3tNZuSfLmJF8eTI848NH672nhVXVQks8m+d+ttX/ZWmtJ3jc4j+cnWTyo/99PYHe/lWT+uP4fTvKHSV6Q5KVJ1lTVokHfP01y+uBcnpXkp0kuGrevv01yWJLfTPIPSa7ojvXaJH+e5KlJbk9y7oRPGmCGCcnAXHVha+17rbV7MhbWXjdo35bk3a21ba21zyS5L8nhVbVPxkLeWa21e1trm5L8lyT/fGc7n2z/SXhWkv+V5OOttXcNjlVJViX51621e1pr9yZ57+D4j2VbknNba9uSfCzJwUk+OKj5W0m+nWTZoO+bk7yztba5tfZAknOSvLqq9k2S1tqlg+0eWresqkbGHetTrbUNrbXtGQvQz9/9rwFgeu077AIAhuR74z5/N2PhM0m2DELcQ36e5MkZC4/7DfqO3+6QXex/sv0n6tSMBfeLx7U9PcmTktw8lpeTJJVknwnsb0tr7VeDz78Y/Pmjcet/kbHzT5LnJPlUVT04bv2vkjyjqn6Ysf/ZeM2gnof6HJxkdPD5h+O2e+h7BZiVXEkG5qpnj/t8aJL/9xj9f5Kxq67P6bb7/uBzm2T/3fXhJP8zyWeq6jfGHesXSZa21g4c/Iy01qY6hH4vySnjjnFga21+a+37Sf4gyWlJfjdjU0wWDrapne8KYHYTkoG56k+qasHgxrN3JvnrR+s8uNp6ZZJzBzfNPSfJv0nyPwZdfpRkQVU9YYL998Rbk9yWZF1VPbG19mDGwvMHquo3k6SqDqmqk6bgWONdnLHzec7gGE+vqtMG6w5I8kCSLRm7qv3eKT42wIwSkoG56iNJrktyZ5I7kkzkJR1/mmTrYJsvDvZx6WDd55J8K8kPq+onE+i/2wY36q1KsjnJ1YMnc/y7jN0M95Wq+sck1yc5fE+P1flgkmuSXFdV9yb5SpIXDdZdnrHpJN/P2Dzmr0zxsQFmVI39txZg7qiqTUn+RWvt+mHXAsDs5EoyAAB0hGSAGTZ40cd9O/m5+LG3nvSxzt7Fsf52qo8F8HhiugUAAHRcSQYAgI6QDAAAnVnxxr2DDz64LVy4cNhlAADwOHfzzTf/pLX29MfqN9SQXFUrk6xcvHhxbrrppmGWAgDAHFBV351Iv6FOt2itrWutrRoZGRlmGQAA8AjmJAMAQEdIBgCAzqy4cQ8AgN2zbdu2bN68Offff/+wS5lV5s+fnwULFmS//fbbre2FZACAvdjmzZtzwAEHZOHChamqYZczK7TWsmXLlmzevDmLFi3arX2YbgEAsBe7//77c9BBBwnI41RVDjrooD26uj7UkFxVK6tq7ejo6DDLAADYqwnIO9rT78Qj4AAAGJobbrghr3jFK4Zdxg7MSQYAeBxZuPraKd3fpvedOqX721uYkwwAwB7ZtGlTjjjiiJx55pl53vOelzPOOCPXX399jjvuuBx22GHZsGFDNmzYkGOPPTbLly/PS17yktx222077Gfr1q154xvfmGOOOSbLly/P1VdfPYSzGSMkAwCwx26//fa87W1vy6233ppbb701H/nIR/LFL34x5513Xt773vfmiCOOyBe+8IVs3Lgx7373u3P22WfvsI9zzz03J554YjZs2JD169fn7W9/e7Zu3TqEs5nj0y1259cRc/VXDgAAj2bRokU56qijkiRLly7Ny172slRVjjrqqGzatCmjo6N5/etfn+985zupqmzbtm2HfVx33XW55pprct555yUZe3LHXXfdlSVLlszouSRzPCQDADA19t9//4c/z5s37+HlefPmZfv27VmzZk1OOOGEfOpTn8qmTZty/PHH77CP1lquuuqqHH744TNV9i6ZbgEAwLQbHR3NIYcckiS57LLLdtrnpJNOygUXXJDWWpJk48aNM1XeDjwnGQCAafeOd7wjZ511VpYvX57t27fvtM+aNWuybdu2HH300Vm6dGnWrFkzw1X+Wj2U1IdpxYoV7aabbprx45qTDADs7W655ZahzNndG+zsu6mqm1trKx5rW9MtAACgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA9sj555+fJUuW5IwzzpiW/Z9zzjkPv6p6pngtNQDA48k5I1O8v8d+6duHPvShXH/99VmwYMHUHnuIvHEPAIDd9uY3vzl33nlnTjnllJx77rl54xvfmGOOOSbLly/P1VdfnWTsNdSnn356Xv7yl2fhwoW58MIL8/73vz/Lly/Pi1/84txzzz1Jkg9/+MN54QtfmGXLluVVr3pVfv7zn+9wvDvuuCMnn3xyXvCCF+SlL31pbr311mk5r6GG5NbautbaqpGRKf4/nserc0Ym9wMAMM0uvvjiPOtZz8r69euzdevWnHjiidmwYUPWr1+ft7/97dm6dWuS5Jvf/GY++clP5sYbb8w73/nOPOlJT8rGjRtz7LHH5vLLL0+SvPKVr8yNN96Yr3/961myZEkuueSSHY63atWqXHDBBbn55ptz3nnn5S1vecu0nJfpFgAATInrrrsu11xzzcPzh++///7cddddSZITTjghBxxwQA444ICMjIxk5cqVSZKjjjoq3/jGN5KMBel3vetd+dnPfpb77rsvJ5100iP2f9999+VLX/pSXvOa1zzc9sADD0zLuQjJAABMidZarrrqqhx++OGPaP/qV7+a/fff/+HlefPmPbw8b968bN++PUly5pln5tOf/nSWLVuWyy67LDfccMMj9vPggw/mwAMPzNe+9rXpPZF4ugUAAFPkpJNOygUXXJDWWpJk48aNk9r+3nvvzTOf+cxs27YtV1xxxQ7rn/KUp2TRokX5+Mc/nmQslH/961/f88J3QkgGAGBKrFmzJtu2bcvRRx+dpUuXZs2aNZPa/j3veU9e9KIX5bjjjssRRxyx0z5XXHFFLrnkkixbtixLly59+ObAqVYPJf1hWrFiRbvppptm/LgLV1876W02ve/UaahkgiZ7M94EHtkCAOzdbrnllixZsmTYZcxKO/tuqurm1tqKx9rWlWQAAOgIyQAA0BGSAQCgIyQDAOzlZsM9ZrPNnn4nXksNALAXmz9/frZs2SIoj9Nay5YtWzJ//vzd3sdQXybSWluXZN2KFSv+eJh1AADsrRYsWJDNmzfnxz/+8bBLmVXmz5+fBQsW7Pb23rgHALAX22+//bJo0aJhl/G4IyTz2Cb7fObEM5oBgL2aG/cAAKDjSvJkeesdAMDjnivJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIYakqtqZVWtHR31LGEAAGaPoYbk1tq61tqqkZHdeO0xAABME9MtAACg47XUQ7Jw9bWT3mbT/GkoBACAHbiSDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANAZakiuqpVVtXZ0dHSYZQAAwCMMNSS31ta11laNjIwMswwAAHgE0y0AAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAICOkAwAAJ19h10AM2/h6msn1X/T/GkqBABglnIlGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI7nJDO7nTMyyf6j01MHADCnuJIMAACdoYbkqlpZVWtHR139AwBg9hhqSG6trWutrRoZmeSv1AEAYBqZbgEAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0PEyEdgVLzIBgDnLlWQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANDZd9gFMLcsXH3tpPpvmj9NhQAAPApXkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgM6Uh+Sqem5VXVJVn5jqfQMAwEyYUEiuqkur6u6q+mbXfnJV3VZVt1fV6iRprd3ZWnvTdBQLAAAzYaJXki9LcvL4hqraJ8lFSU5JcmSS11XVkVNaHQAADMGEQnJr7fNJ7umaj0ly++DK8S+TfCzJaVNcHwAAzLg9mZN8SJLvjVvenOSQqjqoqi5OsryqztrVxlW1qqpuqqqbfvzjH+9BGQAAMLX2neodtta2JHnzBPqtTbI2SVasWNGmug4AANhde3Il+ftJnj1uecGgDQAA9mp7EpJvTHJYVS2qqickeW2Sa6amLAAAGJ6JPgLuo0m+nOTwqtpcVW9qrW1P8tYkf5fkliRXtta+NX2lAgDAzJjQnOTW2ut20f6ZJJ+Z0ooAAGDIpvzGvcmoqpVJVi5evHiYZcDscs7IbmwzOvV1AMAcNuWvpZ6M1tq61tqqkZHdCAUAADBNhhqSAQBgNhKSAQCgIyQDAEBHSAYAgI6QDAAAnaGG5KpaWVVrR0c9vgoAgNnDI+AAAKBjugUAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjuckAwBAx3OSAQCgY7oFAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6+w7z4FW1MsnKxYsXD7MMmFYLV187qf6b5k9TIQDAhHnjHgAAdEy3AACAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAZ6ghuapWVtXa0dHRYZYBAACP4I17AADQMd0CAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANAZakiuqpVVtXZ0dHSYZQAAwCMMNSS31ta11laNjIwMswwAAHgE0y0AAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOjsO8yDV9XKJCsXL148zDLgcWvh6msn1X/T/D+Y/EHOGR3OsXdxXACYCkO9ktxaW9daWzUyMjLMMgAA4BFMtwAAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0Nl3mAevqpVJVi5evHiYZTAHLFx97aS32TR/GgphdjpnZJL9R6enDgBmjaFeSW6trWutrRoZmeRfUAAAMI1MtwAAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdIRkAADoCMkAANARkgEAoCMkAwBAR0gGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCgIyQDAEBHSAYAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6AjJAADQEZIBAKAjJAMAQEdIBgCAjpAMAAAdIRkAADpCMgAAdPad6h1W1W8k+VCSXya5obV2xVQfAwAAptOEriRX1aVVdXdVfbNrP7mqbquq26tq9aD5lUk+0Vr74yS/P8X1AgDAtJvodIvLkpw8vqGq9klyUZJTkhyZ5HVVdWSSBUm+N+j2q6kpEwAAZs6EQnJr7fNJ7umaj0lye2vtztbaL5N8LMlpSTZnLChPeP8AADCb7Mmc5EPy6yvGyVg4flGS85NcWFWnJlm3q42ralWSVUly6KGH7kEZALPDwtXXTqr/pved+rg4NnPHXjXOzhmZZP/R6aljL7VX/bOeJlN+415rbWuSN0yg39oka5NkxYoVbarrAACA3bUn0yG+n+TZ45YXDNoAAGCvtich+cYkh1XVoqp6QpLXJrlmasoCAIDhmegj4D6a5MtJDq+qzVX1ptba9iRvTfJ3SW5JcmVr7VvTVyoAAMyMCc1Jbq29bhftn0nymSmtCAAAhmyoj2irqpVVtXZ01B2lAADMHkMNya21da21VSMjk3xMCwAATCMv+wAAgI6QDAAAHSEZAAA6QjIAAHSEZAAA6FRrbXgHr1qZZGWS/y/Jd2bgkAcn+ckMHIfHD2OG3WHcMFnGDJNlzOy+57TWnv5YnYYakmdaVd3UWlsx7DrYexgz7A7jhskyZpgsY2b6mW4BAAAdIRkAADpzLSSvHXYB7HWMGXaHccNkGTNMljEzzebUnGQAAJiIuXYlGQAAHtOcCMlVdXJV3VZVt1fV6mHXw+xRVZdW1d1V9c1xbU+rqr+vqu8M/nzqoL2q6vzBOPpGVf2T4VXOsFTVs6tqfVV9u6q+VVV/Nmg3btipqppfVRuq6uuDMfPng/ZFVfXVwdj466p6wqB9/8Hy7YP1C4dZP8NTVftU1caq+pvBsjEzgx73Ibmq9klyUZJTkhyZ5HVVdeRwq2IWuSzJyV3b6iSfba0dluSzg+VkbAwdNvhZleQvZqhGZpftSd7WWjsyyYuTlKJXPgAAAu9JREFU/MngvynGDbvyQJITW2vLkjw/yclV9eIk/znJB1pri5P8NMmbBv3flOSng/YPDPoxN/1ZklvGLRszM+hxH5KTHJPk9tbana21Xyb5WJLThlwTs0Rr7fNJ7umaT0vyV4PPf5Xk9HHtl7cxX0lyYFU9c2YqZbZorf2gtfYPg8/3ZuwvsENi3LALg3/29w0W9xv8tCQnJvnEoL0fMw+NpU8keVlV1QyVyyxRVQuSnJrkLwfLFWNmRs2FkHxIku+NW948aINdeUZr7QeDzz9M8ozBZ2OJRxj8SnN5kq/GuOFRDH5t/rUkdyf5+yR3JPlZa237oMv4cfHwmBmsH01y0MxWzCzwX5O8I8mDg+WDYszMqLkQkmG3tbHHv3gEDDuoqicnuSrJv2qt/eP4dcYNvdbar1prz0+yIGO/4TxiyCUxi1XVK5Lc3Vq7edi1zGVzISR/P8mzxy0vGLTBrvzooV+HD/68e9BuLJEkqar9MhaQr2itfXLQbNzwmFprP0uyPsmxGZt6s+9g1fhx8fCYGawfSbJlhktluI5L8vtVtSlj00RPTPLBGDMzai6E5BuTHDa4I/QJSV6b5Joh18Tsdk2S1w8+vz7J1ePa/2jwtIIXJxkd9+t15ojBPL9LktzSWnv/uFXGDTtVVU+vqgMHn5+Y5OUZm8u+PsmrB936MfPQWHp1ks81LzWYU1prZ7XWFrTWFmYst3yutXZGjJkZNSdeJlJVv5exuT37JLm0tXbukEtilqiqjyY5PsnBSX6U5D8k+XSSK5McmuS7Sf5Za+2eQTi6MGNPw/h5kje01m4aRt0MT1X9TpIvJPk/+fVcwbMzNi/ZuGEHVXV0xm6q2idjF6eubK29u6qem7GrhE9LsjHJH7bWHqiq+Un+e8bmu9+T5LWttTuHUz3DVlXHJ/m3rbVXGDMza06EZAAAmIy5MN0CAAAmRUgGAICOkAwAAB0hGQAAOkIyAAB0hGQAAOgIyQAA0BGSAQCg8/8DBrYqnepC3iIAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 864x1728 with 4 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "features = ['photo_cnt', 'photo_like_cnt', 'photo_like_median', 'photo_like_mean']\n",
    "fig, axes = plt.subplots(4, 1, figsize=(12,24))\n",
    "for i in range(4):\n",
    "    axes[i].hist([df[features[i]] [df.male_dummy == 1],df[features[i]] [df.male_dummy == 0]], log=True, bins=20)\n",
    "    axes[i].legend(['male', 'female'], )\n",
    "    axes[i].set_title(features[i])\n",
    "\n",
    "fig.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "Давайте добавим к нашим переменным (описывающим фотографии) количество друзей, количество подарков и пол"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "vars_interesting = features\n",
    "vars_interesting.extend(['male_dummy', 'friends_cnt', 'gifts_cnt'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>photo_cnt</th>\n",
       "      <th>photo_like_cnt</th>\n",
       "      <th>photo_like_median</th>\n",
       "      <th>photo_like_mean</th>\n",
       "      <th>male_dummy</th>\n",
       "      <th>friends_cnt</th>\n",
       "      <th>gifts_cnt</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4.0</td>\n",
       "      <td>309.0</td>\n",
       "      <td>72.0</td>\n",
       "      <td>77.250000</td>\n",
       "      <td>0</td>\n",
       "      <td>298.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>19.0</td>\n",
       "      <td>2894.0</td>\n",
       "      <td>152.0</td>\n",
       "      <td>152.315789</td>\n",
       "      <td>0</td>\n",
       "      <td>408.0</td>\n",
       "      <td>203.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>7.0</td>\n",
       "      <td>2406.0</td>\n",
       "      <td>328.0</td>\n",
       "      <td>343.714286</td>\n",
       "      <td>1</td>\n",
       "      <td>2969.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>50.0</td>\n",
       "      <td>1310.0</td>\n",
       "      <td>25.0</td>\n",
       "      <td>26.200000</td>\n",
       "      <td>0</td>\n",
       "      <td>214.0</td>\n",
       "      <td>24.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>35.0</td>\n",
       "      <td>1375.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>39.285714</td>\n",
       "      <td>1</td>\n",
       "      <td>336.0</td>\n",
       "      <td>227.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   photo_cnt  photo_like_cnt  photo_like_median  photo_like_mean  male_dummy  \\\n",
       "0        4.0           309.0               72.0        77.250000           0   \n",
       "1       19.0          2894.0              152.0       152.315789           0   \n",
       "2        7.0          2406.0              328.0       343.714286           1   \n",
       "3       50.0          1310.0               25.0        26.200000           0   \n",
       "4       35.0          1375.0               19.0        39.285714           1   \n",
       "\n",
       "   friends_cnt  gifts_cnt  \n",
       "0        298.0        NaN  \n",
       "1        408.0      203.0  \n",
       "2       2969.0        NaN  \n",
       "3        214.0       24.0  \n",
       "4        336.0      227.0  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[vars_interesting].head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "__[1]__ Визуализируйте вытащенные признаки попарно. Постройте `pairplot` для всех пар. На диагонали расположите оценки плотностей для признаков. Для этого используйте опцию `diag_kind = 'kde'`. С помощью параметра `hue` разбейте все наблюдения на мужские и женские, чтобы они отражались на картинках разными цветами."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "sns.pairplot(df[vars_interesting], diag_kind=\"kde\", hue='male_dummy', dropna=True);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Давайте для удобства визуализации прологарифмуем наш датасет (предварительно заполнив пропуски нулями или выкинув их)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "import math"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_na_zero = df[vars_interesting].fillna(0)\n",
    "for column in df_na_zero.columns:\n",
    "    if column == 'male_dummy':\n",
    "        continue\n",
    "    df_na_zero[column] = df_na_zero[column].map(lambda x: math.log(1 + x))\n",
    "sns.pairplot(df_na_zero, diag_kind=\"kde\", hue='male_dummy', dropna=True);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_na_drop = df[vars_interesting].dropna()\n",
    "for column in df_na_zero.columns:\n",
    "    if column == 'male_dummy':\n",
    "        continue\n",
    "    df_na_drop[column] = df_na_drop[column].map(lambda x: math.log(1 + x))\n",
    "sns.pairplot(df_na_drop, diag_kind=\"kde\", hue='male_dummy', dropna=True);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "__[2]__ Дайте ответы на соедущие вопросы. Свои мысли пишите прямо здесь! \n",
    "\n",
    "* Между какими признаками скорее всего есть линейная зависимость? Почему вы так решили? \n",
    "\n",
    "__Ответ:__  очевидно, медиана и среднее почти 1:1\n",
    "\n",
    "* Eсть ли тут нелинейные зависимости?\n",
    "\n",
    "__Ответ:__  на логарифмированных графиках почти линейные зависимости между количеством фото и лайков на них, количеством друзей и подарков, значит исходные близки к $y = x^a$\n",
    "\n",
    "* Есть ли объекты-выбросы, которые наблюдаются далеко от сгустков из точек. \n",
    "\n",
    "__Ответ:__ Да есть. __Подойдут любые попытки рассказать о том, что в некоторых облаках точки далеко от основного сгустка.__ \n",
    "\n",
    "* Какие ещё интересные особенности вы видите на картинке?\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "__[1]__ Отлично! Половина домашки осталась позади. Давайте теперь для тех же самых признаков построим матрицу корреляций и провизуализируем её с помощью `heatmap`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.subplots(figsize=(8, 8))\n",
    "sns.heatmap(df[vars_interesting].corr( ), square=True,\n",
    "              annot=True, fmt=\".1f\", linewidths=0.1, cmap=\"RdBu\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.subplots(figsize=(8, 8))\n",
    "sns.heatmap(df_na_zero.corr( ), square=True,\n",
    "              annot=True, fmt=\".1f\", linewidths=0.1, cmap=\"RdBu\");"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "plt.subplots(figsize=(8, 8))\n",
    "sns.heatmap(df_na_drop.corr( ), square=True,\n",
    "              annot=True, fmt=\".1f\", linewidths=0.1, cmap=\"RdBu\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "__[2]__ Между какими переменными корреляция самая высокая? Почему? Она отрицательная или положительная. Прокомментируйте все клетки, где она оказалась $\\ge 0.3$ либо $\\le -0.2$. \n",
    "\n",
    "__Ответ:__ "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "__[1]__ Дамми-переменная это переменная, которая принмает два значения. Либо $1$, если человек обладает закодированным в ней свойством, либо $0$, если не обладает. В нашей таблице все дамми-переменные оканчиваются на суффикc `dummy`.  \n",
    "\n",
    "Возьмите переменную `instagram_dummy`. Она принимает значение $1$, если у пользователя на страничке есть ссылка на инстаграм. Возьмите переменную `male_dummy`. Она примает значение $1$, если пользователь парень. Постройте картинку, на которой будет видно как между собой соотносятся владельцы инстаграмма по полу. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>male_dummy</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>instagram_dummy</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>174</td>\n",
       "      <td>168</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>56</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "male_dummy         0    1\n",
       "instagram_dummy          \n",
       "0                174  168\n",
       "1                 56   27"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dff = df.pivot_table(index='instagram_dummy', columns='male_dummy', values='first_name', aggfunc='count')\n",
    "dff"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x13207b470>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAXoAAAEHCAYAAACgHI2PAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAF2NJREFUeJzt3X2QVfWd5/H3B2lEV+IDdihCkwUUAyIIsSU+FelgXJQ1orvRwTGDCC4zFm5MYhnRCo5YIUVqnHErcZYtZnXA3QDD6LCwjBLRkTFMoh0IyLOR1U5oQhSJIQFFof3uH/d0e2374Xbf29zmx+dVdeue8ztP33u769Onf/ec31VEYGZm6epR7gLMzKxrOejNzBLnoDczS5yD3swscQ56M7PEOejNzBLnoDczS5yD3swscQ56M7PE9Sx3AQBnn312DBo0qNxlmJkdVzZs2PB2RFS2t163CPpBgwaxfv36cpdhZnZckfSrQtZrt+tGUm9JtZJekbRN0pysfaGkNyRtyh6js3ZJ+oGkXZI2S/p8cS/FzMyKUcgZ/fvA+Ig4KKkCWCfpmWzZPRHxZLP1rwGGZo8vAPOzZzMzK4N2z+gj52A2W5E92hrychLwRLbdS8AZkvoXX6qZmXVGQX30kk4CNgDnAn8bES9LugOYK+kB4HlgVkS8DwwAdudtXp+17S1p5WZ2XDty5Aj19fUcPny43KV0e71796aqqoqKiopObV9Q0EdEAzBa0hnAckkXAPcBvwV6AQuAe4GHCj2wpBnADIDPfvazHSzbzI539fX19OnTh0GDBiGp3OV0WxHB/v37qa+vZ/DgwZ3aR4euo4+I3wMvAFdHxN6se+Z94O+Bsdlqe4CBeZtVZW3N97UgIqojorqyst2rg8wsMYcPH6Zv374O+XZIom/fvkX951PIVTeV2Zk8kk4BrgJ2Nva7K/dTuh7Ymm2yEpiSXX1zCXAgItxtY2af4JAvTLHvUyFdN/2BRVk/fQ9gWUSskvQvkioBAZuAv8jWfxqYCOwC3gVuK6pCMzMrSrtBHxGbgTEttI9vZf0AZhZfWjf04OnlriAtDx4odwVmJwSPdWNmSVi7di3XXntt0fupqalJ7k59B72ZWeIc9GbWbdTV1TFs2DCmTp3Keeedxy233MJzzz3H5ZdfztChQ6mtraW2tpZLL72UMWPGcNlll/Hqq69+Yj+HDh1i2rRpjB07ljFjxrBixYpWj/nee+8xefJkhg8fzg033MB7773XtOy0005rmn7yySeZOnUqAFOnTuWOO+7gkksuYciQIaxdu5Zp06YxfPjwpnUat7/nnnsYMWIEX/7yl6mtraWmpoYhQ4awcuVKAMaNG8emTZuatrniiit45ZVXOvsWtshBb2bdyq5du7j77rvZuXMnO3fuZPHixaxbt46HH36Y733vewwbNoyf/OQnbNy4kYceeoj777//E/uYO3cu48ePp7a2lhdeeIF77rmHQ4cOtXi8+fPnc+qpp7Jjxw7mzJnDhg0bCqrznXfe4Wc/+xmPPPII1113Hd/85jfZtm0bW7ZsaQruQ4cOMX78eLZt20afPn34zne+w5o1a1i+fDkPPPAAANOnT2fhwoUA/PKXv+Tw4cNceOGFnXjnWuegN7NuZfDgwYwcOZIePXowYsQIrrzySiQxcuRI6urqOHDgADfeeCMXXHBBU7g29+yzzzJv3jxGjx5NTU0Nhw8f5te//nWLx3vxxRf52te+BsCoUaMYNWpUQXV+5StfaaqrX79+H6u5rq4OgF69enH11VcDMHLkSL74xS9SUVHR9FoAbrzxRlatWsWRI0d4/PHHP/YfQal0i2GKzcwanXzyyU3TPXr0aJrv0aMHR48eZfbs2XzpS19i+fLl1NXVUVNT84l9RARPPfUUn/vc54qqJf/69eY3LOXX1bzmo0ePAlBRUdG0j5ZeC8Cpp57KVVddxYoVK1i2bFnB/1F0hM/ozey4cuDAAQYMGADQ1OXR3IQJE/jhD39I7mpv2LhxY6v7GzduHIsXLwZg69atbN68uWlZv3792LFjBx9++CHLly8v0Sv4pNtvv52vf/3rXHzxxZx55pkl37+D3syOK9/+9re57777GDNmTNNZcXOzZ8/myJEjjBo1ihEjRjB79uxW93fHHXdw8OBBhg8fzgMPPMBFF13UtGzevHlce+21XHbZZfTv33WD8F500UV86lOf4rbbuub+UjX+xSun6urqOC6uW/UNU6XlG6ZOaDt27GD48OHlLqNb+M1vfkNNTQ07d+6kR4+Wz79ber8kbYiI6vb27zN6M7MyeuKJJ/jCF77A3LlzWw35YvnDWDM7Ifz4xz/m3nvv/Vjb4MGDu7TvvRBTpkxhypQpXXoMB72ZnRAmTJjAhAkTyl1GWbjrxswscQ56M7PEOejNzBLnPnozS8agWf9c0v3VzfuPBa23evVq7rrrLhoaGrj99tuZNWtWSesols/ozcyK0NDQwMyZM3nmmWfYvn07S5YsYfv27eUu62Mc9GZmRaitreXcc89lyJAh9OrVi8mTJ7c5LHI5OOjNzIqwZ88eBg4c2DRfVVXFnj17yljRJznozcwS56A3MyvCgAED2L17d9N8fX190+ia3UW7QS+pt6RaSa9I2iZpTtY+WNLLknZJ+gdJvbL2k7P5XdnyQV37EszMyufiiy/mtdde44033uCDDz5g6dKlXHfddeUu62MKubzyfWB8RByUVAGsk/QM8C3gkYhYKul/ANOB+dnzOxFxrqTJwPeBP+mi+s3MmhR6OWQp9ezZk0cffZQJEybQ0NDAtGnTGDFixDGvoy3tBn3kxjE+mM1WZI8AxgN/mrUvAh4kF/STsmmAJ4FHJSm6w3jIZmZdYOLEiUycOLHcZbSqoD56SSdJ2gS8BawB/h/w+4hoHPW/HmjslBoA7AbIlh8A+payaDMzK1xBQR8RDRExGqgCxgLDij2wpBmS1ktav2/fvmJ3Z2ZmrejQVTcR8XvgBeBS4AxJjV0/VUDjhaN7gIEA2fLTgf0t7GtBRFRHRHVlZWUnyzczs/YUctVNpaQzsulTgKuAHeQC/6vZarcCjbeCrczmyZb/i/vnzczKp5CrbvoDiySdRO4Pw7KIWCVpO7BU0neBjcBj2fqPAf9L0i7gd8DkLqjbzMwKVMhVN5uBMS20v06uv755+2HgxpJUZ2ZmRfMwxWaWjgdPL/H+DrS7yrRp01i1ahWf/vSn2bp1a2mPXyIeAsHMrAhTp05l9erV5S6jTQ56M7MijBs3jrPOOqvcZbTJQW9mljgHvZlZ4hz0ZmaJc9CbmSXOl1eaWToKuByy1G6++WbWrl3L22+/TVVVFXPmzGH69OnHvI62OOjNzIqwZMmScpfQLnfdmJklzkFvZpY4B72ZlY0Hti1Mse+Tg97MyqJ3797s37/fYd+OiGD//v307t270/vwh7FmVhZVVVXU19fjb5hrX+/evamqqur09g56MyuLiooKBg8eXO4yTgjuujEzS5yD3swscQ56M7PEOejNzBLnoDczS5yD3swscQ56M7PEtRv0kgZKekHSdknbJN2VtT8oaY+kTdljYt4290naJelVSRO68gWYmVnbCrlh6ihwd0T8QlIfYIOkNdmyRyLi4fyVJZ0PTAZGAJ8BnpN0XkQ0lLJwMzMrTLtn9BGxNyJ+kU3/EdgBDGhjk0nA0oh4PyLeAHYBY0tRrJmZdVyH+uglDQLGAC9nTXdK2izpcUlnZm0DgN15m9XTwh8GSTMkrZe03mNdmJl1nYKDXtJpwFPANyLiD8B84BxgNLAX+OuOHDgiFkREdURUV1ZWdmRTMzPrgIKCXlIFuZD/UUT8E0BEvBkRDRHxIfB3fNQ9swcYmLd5VdZmZmZlUMhVNwIeA3ZExN/ktffPW+0GYGs2vRKYLOlkSYOBoUBt6Uo2M7OOKOSqm8uBPwO2SNqUtd0P3CxpNBBAHfDnABGxTdIyYDu5K3Zm+oobM7PyaTfoI2IdoBYWPd3GNnOBuUXUZWZmJeI7Y83MEuegNzNLnIPezCxxDnozs8Q56M3MEuegNzNLnIPezCxxDnozs8Q56M3MEuegNzNLnIPezCxxDnozs8Q56M3MEuegNzNLnIPezCxxDnozs8Q56M3MEuegNzNLnIPezCxxDnozs8Q56M3MEtdu0EsaKOkFSdslbZN0V9Z+lqQ1kl7Lns/M2iXpB5J2Sdos6fNd/SLMzKx1hZzRHwXujojzgUuAmZLOB2YBz0fEUOD5bB7gGmBo9pgBzC951WZmVrB2gz4i9kbEL7LpPwI7gAHAJGBRttoi4PpsehLwROS8BJwhqX/JKzczs4J0qI9e0iBgDPAy0C8i9maLfgv0y6YHALvzNqvP2prva4ak9ZLW79u3r4Nlm5lZoQoOekmnAU8B34iIP+Qvi4gAoiMHjogFEVEdEdWVlZUd2dTMzDqgoKCXVEEu5H8UEf+UNb/Z2CWTPb+Vte8BBuZtXpW1mZlZGRRy1Y2Ax4AdEfE3eYtWArdm07cCK/Lap2RX31wCHMjr4jEzs2OsZwHrXA78GbBF0qas7X5gHrBM0nTgV8BN2bKngYnALuBd4LaSVmxmZh3SbtBHxDpArSy+soX1A5hZZF1mZlYivjPWzCxxDnozs8Q56M3MEuegNzNLnIPezCxxDnozs8Q56M3MEuegNzNLnIPezCxxDnozs8QVMtaNZQYdXlzuEpJSV+4CzE4QPqM3M0ucg97MLHEOejOzxDnozcwS56A3M0ucg97MLHEOejOzxDnozcwS56A3M0ucg97MLHHtBr2kxyW9JWlrXtuDkvZI2pQ9JuYtu0/SLkmvSprQVYWbmVlhCjmjXwhc3UL7IxExOns8DSDpfGAyMCLb5r9LOqlUxZqZWce1G/QR8SLwuwL3NwlYGhHvR8QbwC5gbBH1mZlZkYrpo79T0uasa+fMrG0AsDtvnfqszczMyqSzQT8fOAcYDewF/rqjO5A0Q9J6Sev37dvXyTLMzKw9nQr6iHgzIhoi4kPg7/ioe2YPMDBv1aqsraV9LIiI6oiorqys7EwZZmZWgE4FvaT+ebM3AI1X5KwEJks6WdJgYChQW1yJZmZWjHa/YUrSEqAGOFtSPfCXQI2k0UCQ+6KgPweIiG2SlgHbgaPAzIho6JrSzcysEO0GfUTc3ELzY22sPxeYW0xRZmZWOr4z1swscQ56M7PEOejNzBLnoDczS5yD3swscQ56M7PEOejNzBLnoDczS5yD3swscQ56M7PEOejNzBLnoDczS5yD3swscQ56M7PEOejNzBLnoDczS5yD3swscQ56M7PEOejNzBLnoDczS5yD3swscQ56M7PEtRv0kh6X9JakrXltZ0laI+m17PnMrF2SfiBpl6TNkj7flcWbmVn7CjmjXwhc3axtFvB8RAwFns/mAa4BhmaPGcD80pRpZmad1W7QR8SLwO+aNU8CFmXTi4Dr89qfiJyXgDMk9S9VsWZm1nGd7aPvFxF7s+nfAv2y6QHA7rz16rO2T5A0Q9J6Sev37dvXyTLMzKw9RX8YGxEBRCe2WxAR1RFRXVlZWWwZZmbWis4G/ZuNXTLZ81tZ+x5gYN56VVmbmZmVSWeDfiVwazZ9K7Air31KdvXNJcCBvC4eMzMrg57trSBpCVADnC2pHvhLYB6wTNJ04FfATdnqTwMTgV3Au8BtXVCzmZl1QLtBHxE3t7LoyhbWDWBmsUWZmVnp+M5YM7PEOejNzBLnoDczS5yD3swsce1+GGtmx4EHTy93BWl58EC5Kygpn9GbmSXOQW9mljgHvZlZ4hz0ZmaJc9CbmSXOQW9mljgHvZlZ4hz0ZmaJc9CbmSXOQW9mljgHvZlZ4hz0ZmaJc9CbmSXOQW9mljgHvZlZ4hz0ZmaJK+qLRyTVAX8EGoCjEVEt6SzgH4BBQB1wU0S8U1yZZtaWQYcXl7uEpNSVu4ASK8UZ/ZciYnREVGfzs4DnI2Io8Hw2b2ZmZdIVXTeTgEXZ9CLg+i44hpmZFajYoA/gWUkbJM3I2vpFxN5s+rdAv5Y2lDRD0npJ6/ft21dkGWZm1ppivxz8iojYI+nTwBpJO/MXRkRIipY2jIgFwAKA6urqFtcxM7PiFXVGHxF7sue3gOXAWOBNSf0Bsue3ii3SzMw6r9NBL+nfSerTOA38B2ArsBK4NVvtVmBFsUWamVnnFdN10w9YLqlxP4sjYrWknwPLJE0HfgXcVHyZZmbWWZ0O+oh4Hbiwhfb9wJXFFGVmZqXjO2PNzBLnoDczS5yD3swscQ56M7PEOejNzBLnoDczS5yD3swscQ56M7PEOejNzBLnoDczS5yD3swscQ56M7PEOejNzBLnoDczS5yD3swscQ56M7PEOejNzBLnoDczS5yD3swscQ56M7PEOejNzBLXZUEv6WpJr0raJWlWVx3HzMza1iVBL+kk4G+Ba4DzgZslnd8VxzIzs7Z11Rn9WGBXRLweER8AS4FJXXQsMzNrQ1cF/QBgd958fdZmZmbHWM9yHVjSDGBGNntQ0qvlqiVBZwNvl7uI9uj75a7AysC/m6X17wtZqauCfg8wMG++KmtrEhELgAVddPwTmqT1EVFd7jrMmvPvZnl0VdfNz4GhkgZL6gVMBlZ20bHMzKwNXXJGHxFHJd0J/Bg4CXg8IrZ1xbHMzKxtXdZHHxFPA0931f6tTe4Ss+7Kv5tloIgodw1mZtaFPASCmVniHPRmZokr23X0VjqShpG787jxprQ9wMqI2FG+qsysu/AZ/XFO0r3khpgQUJs9BCzxYHJmBv4w9rgn6ZfAiIg40qy9F7AtIoaWpzKz1km6LSL+vtx1nCh8Rn/8+xD4TAvt/bNlZt3RnHIXcCJxH/3x7xvA85Je46OB5D4LnAvcWbaq7IQnaXNri4B+x7KWE527bhIgqQe5oaHzP4z9eUQ0lK8qO9FJehOYALzTfBHw04ho6T9R6wI+o09ARHwIvFTuOsyaWQWcFhGbmi+QtPbYl3Pi8hm9mVni/GGsmVniHPRmZolz0NsxI+mnndzu+uPty+UlrZXkL9iwbsFBb8dMRFzWyU2vB0oS9JJ8AYKdcBz0dsxIOpg912RnvE9K2inpR5KULZsnabukzZIelnQZcB3wV5I2STpH0n+R9HNJr0h6StKp2bbnSHpJ0hZJ3212vJ9IWglsz9r+j6QNkrZl31/cVKOkv8ran5M0Nqv1dUnXtfHaTpG0VNIOScuBU5q/7mz6q5IWZtMLJc3Pan49q/PxbB8LO1KTpBcljc7bZp2kCzv/07KkRIQffhyTB3Awe64BDpD7LuEewM+AK4C+wKt8dDXYGdnzQuCrefvpmzf9XeC/ZtOrgJuz6b9odrxDwOC87c7Knk8BtjbuEwjgmmx6OfAsUAFcCGxq47V9i9w3qQGMAo4C1fmvO5v+KrAw73U1jlM0CfgDMDJ7TzYAowutCbgV+G/Z9HnA+nL/vP3oPg+f0Vu51EZEfeTuAdgEDCIX/oeBxyT9J+DdVra9IDtD3wLcAozI2i8F/jGbXtzC8d7Im/+6pFfI3X8wEGgcE+gDYHU2vQX418iNI7Qlq7E144D/DRARm4HW7gpt7v9GRGT7fzMitmTvyba84xVS0z8C10qqAKaR+yNiBrjrxsrn/bzpBqBnRBwld4fvk8C1fBRuzS0E7oyIkeTGTOldwPEONU5IqgG+DFwaERcCG/P2cSQLXsiNFfQ+NN2U1tn+/fybVZrX2vg+fMjH35P847VbU0S8C6wh95/BTcCPOlmrJchBb92GpNOA0yP3fcPfJNc1AfBHoE/eqn2AvdnZ6y157S8B/zmbntzGoU4H3omId7Ox/C8pQfkvAn8KIOkCct03jd6UNDwbquKGEhyrNf8T+AG54S+aDztgJzAHvXUnfYBV2WBY68j1e0OuH/seSRslnQPMBl4G/g3Ymbf9N4BvZdufS64rqCWrgZ6SdgDzKM3wEfOB07J9PkSuj73RLHKfH/wU2FuCY7UoIjaQ6+f38L/2MR4CwZKRXX3zXkSEpMnkPpidVO66jhVJnwHWAsOybh0zwIOaWVouAh7NLtX8PbkPJU8IkqYAc4FvOeStOZ/Rm3WApAnA95s1vxERXdn3blYUB72ZWeL8YayZWeIc9GZmiXPQm5klzkFvZpY4B72ZWeL+P1ttIPoci5rCAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "dff.plot(kind='bar', stacked=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Правда ли, что девушки чаще указыают наличие инстаграмма на своей страничке?\n",
    "\n",
    "__Ответ:__  Да. Во втором столбике синяя фигня толще. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "__[1]__  Категориальная переменная обычно принимает значения из какого-то фиксированного множества. Например, переменная `political_cat` описывает к какой категории относятся политические взгляды юзера. Постройте для этой переменной столбиковую диаграмму. Разбиритесь по [документации](http://seaborn.pydata.org/generated/seaborn.countplot.html#seaborn.countplot) как сделать у столбиков горизонтальное расположение. Можно ли сделать исходя из картинки вывод, что в вышке одни либералы? Почему? "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAWAAAAFuCAYAAAC/a8I8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvnQurowAAIABJREFUeJzt3XucVWXd9/HPFwRFwErRVMoGDQ+JCjmYeShM49YixTvJysPjITUPPPoUdrg1xbK673ysnrJSNMW01DR73UbdoikokiIjJ/GUWmSkhUgeBg8o/J4/1jW22G5m9oY9XMPM9/16zWvWutZa1/qtQb/72tfee21FBGZmtv71yl2AmVlP5QA2M8vEAWxmlokD2MwsEwewmVkmDmAzs0wcwGZmmTiAzcwycQCbmWXiADYOPvjgAPzjH/807qcmDmBj6dKluUsw65EcwGZmmTiAzcwycQCbmWUi347S+m89JHY+5oLcZZh1eQ9cdGytu6qWnTwCNjPLxAFsZpaJA9jMLBMHsJlZJg5gM7NMHMBmZpk4gM3MMnEAm5ll4gA2M8vEAWxmlokD2MwsEwewmVkmDmAzs0wcwGZmmTiAzcwycQCbmWXiADYzy8QBbGaWiQPYzCyTHh/Akr4u6azS+jclnSkpJB2c2t4h6RVJE9P6DpJulfSApBmSdk7tkyVdKqlF0h8ljUntvSVdJGm2pAWSTkntoyRNKZ17Qukc0yU1l7a1VjsmtQ2StKi9c5lZ19PjAxi4EjgWQFIv4NPAYmBOWzvwWWB+6ZhJwPiI2BOYAPy4tK0J2Av4OHCppE2AE4EXImIkMBI4SdKQTrqems4l6eT0QNHyxssvdVIpZtaejXIXkFtELJL0nKQRwDuBucBzwDPAxpI2Bw4FbgH6ShoA7APcKL35xacbl7r8ZUSsAh6X9CdgZ2A0sLukI9I+bwOGAiuA/SXNS+1bApeX+vq5pFfScr9Se9sxAfww1dZmTef6c8V1T6J4IKH/1kP81dhmGfT4AE6uAI4DtqYYEbf5BfBfwGMUYdmX4lnD8xExfA19VYZZUHxF9fiImFreIGkUMCMi2qYqJgADSrscFREtaVtrqX1GRIyRNAh4FPh9udtq5zKzrsdTEIVfAwdTPGUvB9dvgBGUQjkiXgT+LGkcgAp7lI4ZJ6mXpB2A7SnCeypwqqQ+6ZgdJfVvQN0vAW8AvUttnXUuM2swj4CBiFghaRrFyHZl29RCRKwAmgEkHVQ65CjgJ5LOBfoA1/OvOeKngPuBzYDPR8Srkq6gmBueo6LzZ4Gx61DyPpLuAfoD36MI4jaNPpeZdRJFePovvfg2BxgXEY+vQz+TgSkRcVOjalsf+m89JHY+5oLcZZh1eQ9cdGzHOxXU8S6egkDS+4AngDvWJXzNzOrV46cgIuJhirnaRvR1XCP6MbOeocePgM3McnEAm5ll4gA2M8vEAWxmlokD2MwsEwewmVkmDmAzs0wcwGZmmTiAzcwycQCbmWXiADYzy8QBbGaWiQPYzCwTB7CZWSa+IbvR3NwcLS0tucsw6058Q3Yzs67MAWxmlokD2MwsEwewmVkmDmAzs0wcwGZmmTiAzcwycQCbmWXiADYzy2Sj3AVYfiueeYinvr5b7jLatd15D+YuwazhPAI2M8vEAWxmlokD2MwsEwewmVkmDmAzs0wcwGZmmTiAzcwycQCbmWXiADYzy8QBbGaWiQPYzCwTB7CZWSYOYDOzTBzAZmaZOIDNzDJxAJuZZeIANjPLxAFsZpaJA9jMLJNODWBJTZIWpuVdJM2X9G5JX5C0MP2cVdo3JH0+rfeW9DdJk9P6ZEmLJfVO66em/Zskfb2tn7Ttm5LOlDRK0pRS+wRJE9PydEnNaflCSa2l/b4s6cFU739K2l/SPEkPS3olLc9L+y6SNCgtX1u63omSJlT5O/SWdJGk2ZIWSDplTect15mOu0XS8TXUf3ap/wsa8W9pZo23Xr6UU9Jg4Drgs8BWwPHABwABsyTdBfwTeAIYC1wKHAz8taKrvwH/BvwOOCztD3AlcDPwfUm9gE8DewEdftOkpK2AA0vrh6S+PxARL0vaPCKWAcMlNQFTImJ4lX52A4aVmlal66t0IvBCRIyUtDEwU9JtwM6V56047jLgvoi4qoP6RwND0/ULuEXShyLi7orjTgZOBhj8tj7V/zhm1qnWxxTEAOBW4K6IeAjYD/h1RCyPiFaK4Nw/7fsa8ISkXYFjgGsq+roGOEbSMODxtD8RsQh4TtIIYDQwNyKeS8e0jV7nAf+nSn1fA75VWj8IuCoiXk59L6vxOi8Ezi+tLwZGVNlvNHBsqmcWsAVFYLZ33onAocDFNdQ/Ov3MBeZQBPvQyoMiYlJENEdE8+b9e9dyfWbWYOsjgN9NERAHSNqlhv2vAr5EMTr/R8W2vwN9gLPTfmVXAMdRjK6vLLXPiIjhadT6vYpjmoBhEfGbGupqzz5AKzC/1HYD0C9NPfyu1C5gfFtNETEkIm7roP/XKEbA51S0N/HW+gV8u9T/eyPip2txTWbWydZHAD8SEdcB4ylC5B5grKRNJfUHDgdmtO0cEQ9QTFNUBmybq4CtImJORfuvKaYtRgJTa6ztfFYftQLcDhwvaVOAKlMB1UwEzis3pBH+4RExDPhYadNU4FRJfVL/O6a/Q3vn/TbFCPuw9OygvfqnAidIGpD6GZymKcysi1kvc8AAEXGXpEcp5n4nA/enTVdExNw0v9q27yEAko6o0s9vgd9WaV8haRrwfESsrLGsxZVzoxFxq6ThQIukFRSj1//ooJ9ZEfFk+RracQXFyHWOJAHPAmM7Om9EvCbpNGCSpLYpm2r135aeadxbdE8rcDSwpIbazGw9UkTkrqEh0otvc4BxEfF47no2JLsP7hdTTnlv7jLatd15D+Yuwawe1V6Af4tu8T5gSe+jeEfEHQ5fM9tQrLcpiM4UEQ8D2+euw8ysHt1iBGxmtiFyAJuZZeIANjPLxAFsZpaJA9jMLBMHsJlZJg5gM7NMHMBmZpk4gM3MMnEAm5ll4gA2M8vEAWxmlokD2MwsEwewmVkm3eJ2lLZu+m6zK9ud15K7DLMexyNgM7NMHMBmZpk4gM3MMnEAm5ll4gA2M8vEAWxmlokD2MwsEwewmVkmDmAzs0wUEblrsMwGbDcg9jh7j9xltGvm+Jm5SzCrh2rZySNgM7NMHMBmZpk4gM3MMnEAm5ll4gA2M8vEAWxmlokD2MwsEwewmVkmDmAzs0wcwGZmmTiAzcwycQCbmWXiADYzy8QBbGaWiQPYzCwTB7CZWSYOYDOzTDaoAJa0m6RfSbpf0mxJvXPXZGa2tjbKXUCtJG0FXA58PiLm5a7HzGxdZR0BS2qS9IqkeZKeknSJpOmSmtP2CyW1pt2PAFYB10laKOmsUj9Hp1HxPEmXtY2MJbVK+p6khyTdIWlLSfun/R4unXte2v+8NLJeKGmSJKX26ZL+X9p3oaS9UvtESRPS8kGSolT7TyS1pHNfUKp1kaRBaXmQpEVpubek/5v6XyBpfHl/SQMkzZQ0uko/10paWOrnonQdCySd0hn/dma27moOYEln1tJWp97A4xExHDivou+tgANLTVsCmwHNwN7ASZJGSNoFOBLYN/WzEjgqHdMfaImIXYG7gPMjYkba72PAkxExPK0DXBIRIyNiGNAPGFM6/6Zpv9OAK6tcy3nAE6X1cyKiGdgd+LCk3Tv4W5wMNAHDI2J34OelbX2AG4EfR8Rt5YMk7QYMKzWdCLwQESOBkenvNKTyZJJOTg8QLa+3vt5BaWbWGeoZAf+vKm3HreP5+wGvrmHb14BvldYF3BwRyyOiFbgZ2J8ipPcEZqeR7IHA9umYVcANaflaYL8O6jlA0ixJDwIfAXYtbbsOICLuBjaT9PY3C5M+CcwG/lba/1OS5gBzUz/vK22blmqdVmo7CLgsIt5I51lW2nY5sE1ElEO5zYXA+aX10cCxqf9ZwBbA0MqDImJSRDRHRHOfAX2qdGtmna3DOWBJnwE+CwyRdEtp00BgWfWjarYt8HSV9iZgWESMT7MAAC8Cb6+yr4CrI+KrNZwv1rRB0ibAj4HmiPirpInAJu0c27beGzibYrR8U+prCDABGBkR/5Q0uaKvAyJiaZpCaKmh7seB5ySdEBHl0fc+QCswv3wpwPiImFpDv2aWUS0j4D8AFwOPpt9tP18E/m0dzz8OmFml/XxWH9VBMZo7XNKmkvoDhwMzgDuAI9KUBZI2l/SedEwvirljKB5E7mmnlraAXCppQOm4Nkem/vejeIr/Qmo/GvhdRCwt7bsZsBx4QdI7gUPaOW+b24FTJG3Udh2lbd8EvgB8KfXXZiIVUzfAVOBUSX1SPzumv5eZdTEdjoAj4i/AX4APNvLEkr5DMUf7oyqbF6en+uU6Zkq6EXiAYp738oiYm/o6F7hNUi/gdeD0VPNyYK+0fQkpRKuJiOclXQ4sBP5OMaVQ9qqkuRTzsSeU2t8JfLeir/lp30eBv1L9QabSFcCOwAJJr1NMO1xS6vM5SV8Hfgh8KjXPiognJTVV9NMEzEkvIj4LjK3h/Ga2nilijc/KV99R2pvif/5dgL4UT72XR8RmnVfeupHUGhEDGtDPdGBCRNQyXbDBGbDdgNjj7D1yl9GumeNreQwz6zLU8S71vQh3CfAZivnIfsDnqD56NTOzGtT1PuCIeALoHRErI+Iq4ODOKasxGjH6Tf2M6q6jXzPLp55Pwr0sqS8wL83fPsMG9lFmM7OupJ4APSbtfwbFi1vvBj7ZGUWZmfUE9YyAlwIrIuJV4IL0cd+NO6csM7Pur54R8B3ApqX1fsDvG1uOmVnPUU8Ab5I+AgxAWt60nf3NzKwd9QTwcknvb1uRtCfwSuNLMjPrGeqZAz4LuFHS0xRvMt6adj5ZZmZm7as5gCNitqSdgZ1S02MR8eZ9DCV9NCJub3SBZmbdVb0fxHg9Ihamn8qbyP5XA+syM+v2GvlBipo++2xmZoVGBnBtd/UxMzPAHyU2M8umkQG8qIF9mZl1e7V8JdG/t7c9Im5Ov9vdz8zMVlfL29A+0c62oPhyTNuA7bzVzr7huVkGtXwl0fHroxAzs56mnk/CIenjFF+x/uY3/EbE1xtdlJlZT1Dzi3CSLqX46PF4ivf8jgPe0+5BZma2RvW8C2KfiDgW+GdEXEDxLck7dk5ZZmbdXz0B3Hbns5clbUvx9e/bNL4kM7OeoZ454CmS3g5cBMyheAfEFZ1SlZlZD1DP3dC+kRZ/JWkKxQ3aX+icsszMur96XoQ7PY2AiYjXgF6STuu0yszMurl65oBPiojn21Yi4p/ASY0vycysZ6hnDri3JEVEAKRvRe7bOWXZ+vTSY49x14c+nLuMdn347rtyl2DWcPUE8K3ADZIuS+unpDYzM1sL9QTwlylC99S0fjt+F4SZ2Vqr510Qq4CfpB8zM1tHtdyO8pcR8SlJD1LlWy8iYvdOqczMrJurZQR8Zvo9pjMLMTPraTp8G1pEPJMWT4uIv5R/AL8P2MxsLdXzPuCPVmk7pFGFmJn1NLXMAZ9KMdLdXtKC0qaBgL9GwcxsLdUyB/wL4H+AbwNfKbW/FBHLOqUqM7MeoJYAjohYJOn0yg2SNncIm5mtnVpHwGOAByjehqbStgC274S6zMy6vVq+lHNM+j2k88sxM+s5ankR7v3tbY+IOY0rx8ys56hlCuLidrYF8JEG1WJm1qPUMgVxwPooxMysp6n5ZjyS+lDcCe1DqWk6cFlEvN4JdZmZdXv13I7yJ0Af4Mdp/ZjU9rlGF2Vm1hPUE8AjI2KP0vqdkuY3uiAzs56inntBrJS0Q9uKpO2BlY0vycysZ6hnBHw2ME3Sn9J6E3B8wysyM+sh6hkBzwQuA1YBy9LyvbUeLOlYSQskzZd0jaQmSXemtjskbZf2myzpiNJxCyU1peXpkprT8oWSWtPygNTHHEkPSjqsdHyTpFckzZP0lKRL2usrrX859TNf0n9K2j8d/3Cpr3lp30WSBqXlayUtTMsTJU0o1dDW3lvSRZJmp2s/ZU3nLdeZjrtF0vE11H92qf8L1vDvcbKkFkktL7zu11HNcqhnBPwz4EXgG2n9s8A1wLiODpS0K3AusE9ELJW0OXA1cHVEXC3pBOAHwNhaCpG0FXBgqelV4PCIeDGF4X2Sbknf4NwbeDwihks6Dmhury9JhwCHAR+IiJdL97sYnh4IpkTE8Co17QYMKzWtYvWPbbc5EXghIkZK2hiYKek2YOfK81YcdxlwX0Rc1UH9o4GhwF7p/LdI+lBE3F0+LiImAZMAdho48C3fdGJmna+eEfCwiPhcRExLPycBu9Z47EeAGyNiKUAKtA9S3GcCiiDfr45avgZ8q7Qu4Fvpdpm/BwYD70zb+lEEdK19HQRcFREvl2qtxYXA+aX1xcCIKvuNBo5NI+hZwBYUgdneeScCh1L9QzGV9Y9OP3OBORTBPrTGazCz9aieAJ4jae+2FUkfAFoaX1KHmigeDH5TajsK2BLYM41O/wFskrZtCzxdR19rYx+gFSi/K+QGoF+aevhdqV3A+IgYnn6GRMRtHfT/GsUI+JyK9ibeWr+Ab5f6f29E/HQtrsnMOlk9Abwn8Ic057mIYv53ZJqzXND+odwJjJO0BRS3sQT+AHw6bT8KmFFjHeez+kgT4G3Akoh4XdIBwHtK28ax5hvHV+vrduB4SZuWau3IROC8ckNELI+IwyNiGPCx0qapwKnpgy1I2lFS/w7O+22KEfZhaTqnvfqnAidIGpD6GZymKcysi6lnDvjgtT1JRDwk6ZvAXZJWUjw9Hg9cJels4FlWf0fFNySdlZaHAD8EPpHWF1fOZwI/B36j4pubW4BHASR9B+gP/GgNpb2lr4i4VdJwoEXSCorR6390cImzIuLJthcLO3AFxch1jiRRXPvYjs4bEa9JOg2YJGn/duq/TdIuwL1F97QCRwNLaqjNzNYjFa9TdW2SpkfEqNx1dFc7DRwYk0a0e9O77D589125SzCrR7UX4N+inimInDyHaWbdzgYRwBFxTe4azMwabYMIYDOz7sgBbGaWiQPYzCwTB7CZWSYOYDOzTBzAZmaZOIDNzDJxAJuZZeIANjPLxAFsZpaJA9jMLBMHsJlZJg5gM7NM6rkhu3VTA3fayffbNcvAI2Azs0wcwGZmmTiAzcwycQCbmWXiADYzy8QBbGaWiQPYzCwTB7CZWSYOYDOzTPxJOGPJ4he45Iu/yV3GBu2Miz+RuwTbAHkEbGaWiQPYzCwTB7CZWSYOYDOzTBzAZmaZOIDNzDJxAJuZZeIANjPLxAFsZpaJA9jMLBMHsJlZJg5gM7NMHMBmZpk4gM3MMnEAm5ll4gA2M8vEAWxmlokDGJC0m6RfSbpf0mxJvXPXZGbdX4//SiJJWwGXA5+PiHm56zGznqNbjoAlNUl6RdI8SU9JukTSdEnNafuFklrT7kcAq4DrJC2UdFapn6PTqHiepMvaRsaSWiV9T9JDku6QtGVqPymNoOenEfWmpb4mS/pz6muFpEGSRkmaUqX+4yRdUlq/RNJxaXmRpEEV+0+RNCotj5Z0r6Q5km6UNKAhf1Qza7huGcBAb+DxiBgOnFfekEa8B5aatgQ2A5qBvYGTJI2QtAtwJLBv6mclcFQ6pj/QEhG7AncB56f2myNiZETsATwCnFhR0xdTX0837lJXu7ZBwLnAQRHxfqAF+MIa9j1ZUoukltaXX+iMcsysA911CqIf8Ooatn0N+BZwXVoXRXAuB5B0M7A/xah4T2C2pLY+l6RjVgE3pOVrgZvT8jBJFwJvBwYAU2uoaX9J84AAfhgRV6b2IyXtl5YHU4Rpm2mSVgEPAieV2vcG3gfMTDX3Be6t9keIiEnAJIDtth4a1fYxs87VXQN4W6qPMpuAYRExPgUUwIsUgVlJwNUR8dUaztcWYJOBsRExP00ZjKqhphkRMSaNXh+VdH1qvyEizoBiCqLimAOA54CfAcdU1Hx7RHymhprNLLPuOgUxDphZpf18/jVd0GYWcLikTSX1Bw4HZgB3AEekKQskbS7pPemYXhRzxwCfBe5JywOBZyT14V/TFUh6L0X4P9xOzS8Bb1BMVXQoIgJYRjHKbXMfsG86H5L6S9qxlv7MbP3rdiNgSd+hmKP9UZXNiyPi7nJDRMyUdCPwAMU87+URMTf1dS5wm6RewOvA6cBfgOXAXmn7Eoq5YiimN2YBz6bfAyVtC/w3cHJErKhS0z6S7kk1fy8iXiqNztdkSpqCaKWY4z44XcuzaeR9naSN077nAn/sqEMzW/9UDKSsHpJaI6LbvLtgu62HxpeO+m7uMjZoZ1z8idwlWNfS4SgKuu8UhJlZl+cAXgvdafRrZvk4gM3MMnEAm5ll4gA2M8vEAWxmlokD2MwsEwewmVkmDmAzs0wcwGZmmTiAzcwycQCbmWXiADYzy8QBbGaWiQPYzCwTB7CZWSa+IbvR3NwcLS0tHe9oZrXyDdnNzLoyB7CZWSYOYDOzTBzAZmaZOIDNzDJxAJuZZeIANjPLxAFsZpaJA9jMLBN/Es4YvMU74rRDDsxdxgbtnGtvyl2CdS3+JJyZWVfmADYzy8QBbGaWiQPYzCwTB7CZWSYOYDOzTBzAZmaZOIDNzDJxAJuZZeIANjPLxAFsZpaJA9jMLBMHsJlZJg5gM7NMHMBmZpk4gM3MMnEAm5ll0iMCWNIiSYPa2X6cpEvWZ01mZj0igM3MuqL1HsCSmiQtTMt9JP1J0iWSJks6orTfQklNaXmQpBWS5kl6QtKU1D6qtLy5pOclTUjrB0n6o6SbgL7AxZIekXR66dy/lDQXOBwYJuk+STMlbZn22UHSrZIekDRD0s6pfbKkSyW1pHOMSe29JV0kabakBZJOqawzrU+QNFHS/umaHpb0Slqel/Z5c9Sern9RWn7LaF1Ss6Tpabm/pCsl3S9prqTDGvVvZ2aNlXsEfDLQWsN+vYHFETEc+Nwa9vkq8FRp/UfAx4EzgQHA1UAzcFYK2COBVyNiBPAkEMA+wA3AV1Ifk4DxEbEnMAH4can/JmCvdI5LJW0CnAi8EBEjgZHASZKGrOmiImJGuqaPAU9GxPC0vi7OAe6MiL2AA4CLJPWv3EnSyekBpGX5q6+t4ynNbG1slOvEKRSOpwi1Yan5IknnpuUdSrsPAJa109dgYG/g12n9HUCfiHg8rS8DFkTE8jTC3IMiIH+fulgA9I2IVZLuAC6TNIAikG+U3vyC041Lp/1lRKwCHpf0J2BnYDSwe2kk/zZgKLAC2L9tdAtsCVze0d8ImCZpJcUDUNmRkvYDXgcuAP5e2jYaOLTtmQCwCbAd8Ei5g4iYRPEAw+At3uGvxjbLIFsAU4xMJ1GEU5uzI+ImKKYgSu1DgMXt9HU+8A2KwISOvxJaHewjimcHz7czIq0MrUjHjY+Iqat1Jo0CZkRE21TFBIoHlY4cEBFL01RES6n9hog4Q9JQYDpQnmYQ8MmIeKyG/s0so1xTEG8DxgJX1rj/OGDKGrbtADRFxG1tDRGxDFgp6b1pdLw5xci0PzCCYsTbAhyUDtkd2FVSL+BAYHZEvAj8WdI4ABX2KNckqZekHYDtgceAqcCpkvqkY3as9vS/gZbx1gfRqcB4pWG7pBGdeH4zWwe5RsDvAiZExBulp/dVSTqNYq74w5LOoBg5binpUOBFiqf+x1c5dDzwP8A8innmYymmO34YEf+QdD0wNk0LLErH/IFiJHt4Wj8K+EmaFukDXA/MT9ueAu4HNgM+HxGvSrqCYm54TgrAZykeaBrt3yUNp/hbnF2x7RvA94EF6QHlz8CYTqjBzNaRIrr29J+kicD0iJheahsDDIqIyTX2sQhojoila9h+XNp+Ro39TQamtE2XbOgGb/GOOO2QA3OXsUE759pu8Z+CNU5H06BA3jngWt0ELKlom8PqL4iZmW1wunwAR8TCKm1P19lHUwfbJwOT6+jvuHrOb2ZWTe73AZuZ9VgOYDOzTBzAZmaZOIDNzDJxAJuZZeIANjPLxAFsZpaJA9jMLBMHsJlZJg5gM7NMHMBmZpk4gM3MMnEAm5ll4gA2M8uky9+Q3Tpfc3NztLS0dLyjmdWqphuyewRsZpaJA9jMLBMHsJlZJg5gM7NM/CKcIekl4LHcdXRgEFD1W627iK5eH3T9Grt6fVB7jUsj4uCOduryX8pp68VjEdGcu4j2SGrpyjV29fqg69fY1euDxtfoKQgzs0wcwGZmmTiADWBS7gJq0NVr7Or1QdevsavXBw2u0S/CmZll4hGwmVkmDmAzs0wcwD2cpIMlPSbpCUlfyV1PJUlXSloiaWHuWqqR9G5J0yQ9LOkhSWfmrqlM0iaS7pc0P9V3Qe6a1kRSb0lzJU3JXUs1khZJelDSPEkNuXuV54B7MEm9gT8CHwUWA7OBz0TEw1kLK5H0IaAV+FlEDMtdTyVJ2wDbRMQcSQOBB4CxXeVvKElA/4holdQHuAc4MyLuy1zaW0j6AtAMbBYRY3LXU0nSIqA5Ihr2YRGPgHu2vYAnIuJPEbECuB44LHNNq4mIu4FluetYk4h4JiLmpOWXgEeAwXmr+pcotKbVPumny426JL0L+DhwRe5a1icHcM82GPhraX0xXSg8NjSSmoARwKy8lawuPbWfBywBbo+ILlVf8n3gS8Cq3IW0I4DbJD0g6eRGdOgANmsASQOAXwFnRcSLuespi4iVETEceBewl6QuNZUjaQywJCIeyF1LB/aLiPcDhwCnp+mxdeIA7tn+Bry7tP6u1GZ1SHOrvwJ+HhE3565nTSLieWAa0OFNYtazfYFD0xzr9cBHJF2bt6S3ioi/pd9LgF9TTOGtEwdwzzYbGCppiKS+wKeBWzLXtEFJL3L9FHgkIr6bu55KkraU9Pa03I/iBddH81a1uoj4akS8KyKaKP4bvDMijs5c1mok9U8vsiKpPzAaWOd35jiAe7CIeAM4A5hK8eLRLyPiobxVrU7SdcC9wE6SFks6MXdNFfYFjqEYtc1LPx/LXVTJNsA0SQsoHnBvj4gu+TavLu6dwD2S5gP3A7+NiFvXtVO/Dc3MLBOPgM3MMnEAm5ll4gA2M8vEAWxmloltmRmVAAABXUlEQVQD2MwsEwewWTck6SxJm+auw9rnt6GZdUOdcecuazyPgM0ykXSspAXpXr3XSGqSdGdqu0PSdmm/yZKOKB3Xmn6PkjRd0k2SHpX0cxX+N7AtxQcwpuW5OqvFRrkLMOuJJO0KnAvsExFLJW0OXA1cHRFXSzoB+AEwtoOuRgC7Ak8DM4F9I+IH6d66B3gE3LV5BGyWx0eAG9sCMiKWAR8EfpG2XwPsV0M/90fE4ohYBcwDmjqhVuskDmCzru8N0v+rknoBfUvbXistr8TPajcoDmCzPO4ExknaAiBNQfyB4m5gAEcBM9LyImDPtHwoxbdadOQlYGCjirXO4UdLswwi4iFJ3wTukrQSmAuMB66SdDbwLHB82v1y4L/TnbhuBZbXcIpJwK2Sno6IAxp/BdYIfhuamVkmnoIwM8vEAWxmlokD2MwsEwewmVkmDmAzs0wcwGZmmTiAzcwy+f+M+j0I08w/zwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.catplot(y = 'political_cat', data=df, kind='count');"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "как политические взгляды распределены по полам?  \n",
    "__Ответ:__  Можно визуализировать, но выводы делать нельзя, мало данных"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>male_dummy</th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>political_cat</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>индифферентные</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>коммунистические</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>либеральные</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>либертарианские</th>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>социалистические</th>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>умеренные</th>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "male_dummy          0    1\n",
       "political_cat             \n",
       "индифферентные    1.0  1.0\n",
       "коммунистические  1.0  NaN\n",
       "либеральные       1.0  1.0\n",
       "либертарианские   NaN  1.0\n",
       "социалистические  NaN  1.0\n",
       "умеренные         2.0  3.0"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dff = df.pivot_table(index='political_cat', columns='male_dummy', values='first_name', aggfunc='count')\n",
    "dff"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x136c336d8>"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "dff.plot(kind='bar', stacked=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "__[n]  Удиви нас. Попробуй найти в данных какую-то классную особенность и визуализируйте её. Если у тебя это получится, мы поставим дополнительные баллы.__ Если вы найдёте полную фигню (сколько всего друзей у Маши или типа того), баллов не будет. Найденный факт реально должен выносить мозг и сносить крышу."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# ваш код "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Настрадался? Выскажи всё, что думаешь обо всём этом [в анонимке по четвёртой домашке.](https://docs.google.com/forms/d/e/1FAIpQLSeumKE6K8fB7xZsXneN1KjkV9y_gmLl6wGG7MDiY84HWBUsNg/viewform) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " "
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}