{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<center><img src=\"img/logo_hse_black.jpg\"></center>\n",
    "\n",
    "<h1><center>Методы машинного обучения</center></h1>\n",
    "<h2><center>Семинар: введение</center></h2>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.style.use('ggplot')\n",
    "plt.rcParams['figure.figsize'] = (12,8)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Вспоминаем Pandas"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Удотребление ЛСД и успеваемость"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "В 1968 году была [опубликована](http://www.ncbi.nlm.nih.gov/pubmed/5676802) статья под интригующем названием Correlation of Performance Test Scores with Tissue Concentration of Lysergic Acid Diethylamide in Human Subjects.\n",
    "\n",
    "К статье приложен небольшой набор [данных](https://www.dropbox.com/s/ui14yeeckbc6z7c/drugs-and-math.csv?dl=0), состоящий из 7 наблюдений"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_csv('./data/drugs-and-math.csv', \n",
    "                 index_col=0, \n",
    "                 sep=',')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Таблица уже отсортирована по колонке Drugs - отсортируем по колонке Score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.sort_values('Score', \n",
    "                    ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>Drugs</th>\n",
       "      <th>Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.17</td>\n",
       "      <td>78.93</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3.26</td>\n",
       "      <td>67.47</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2.97</td>\n",
       "      <td>58.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5.83</td>\n",
       "      <td>45.65</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4.69</td>\n",
       "      <td>37.47</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Drugs  Score\n",
       "0   1.17  78.93\n",
       "2   3.26  67.47\n",
       "1   2.97  58.20\n",
       "4   5.83  45.65\n",
       "3   4.69  37.47"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Визуализируйте данные с помощью scatter plot. Видна ли тенденция, как можно ее измерить?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x11d442d30>"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df.plot(x='Drugs', y='Score', kind='scatter')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>Drugs</th>\n",
       "      <th>Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Drugs</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.936928</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Score</th>\n",
       "      <td>-0.936928</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          Drugs     Score\n",
       "Drugs  1.000000 -0.936928\n",
       "Score -0.936928  1.000000"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.corr()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Рождаемость в США"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Загрузите два набора данных с информацией о рождаемости в США: [Набор 1](https://www.dropbox.com/s/4v743y3e25lz0an/US_births_1994-2003_CDC_NCHS.csv?dl=0), [Набор 2](https://www.dropbox.com/s/3aoulbiuomamay6/US_births_2000-2014_SSA.csv?dl=0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "df1 = pd.read_csv('./data/US_births_1994-2003_CDC_NCHS.csv')\n",
    "df2 = pd.read_csv('./data/US_births_2000-2014_SSA.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "`Pandas` обладает мощным функционалом для работы с временными рядами. Самое главное, пожалуй правильно задать столбец с датой в DataFrame. Вернемся к данным:"
   ]
  },
  {
   "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>year</th>\n",
       "      <th>month</th>\n",
       "      <th>date_of_month</th>\n",
       "      <th>day_of_week</th>\n",
       "      <th>births</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1994</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>8096</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1994</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>7772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1994</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>10142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1994</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>11248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1994</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>11053</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   year  month  date_of_month  day_of_week  births\n",
       "0  1994      1              1            6    8096\n",
       "1  1994      1              2            7    7772\n",
       "2  1994      1              3            1   10142\n",
       "3  1994      1              4            2   11248\n",
       "4  1994      1              5            3   11053"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Функция `pd.to_datetime()` преобразовывать сырые данные с специальный тип Timestamp. Делать это можно одним из нескольких способов:\n",
    "* Задан столбец, в котором дата записана в виде строчки произвольного формата (например 2016-03-01, 01032016, 01.03.16 ...)\n",
    "    \n",
    "    В этом случае вам надо указать в `pd.to_datetime()` [формат](http://strftime.org/) даты\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2016-03-01 00:00:00\n",
      "2016-03-01 00:00:00\n",
      "2016-03-01 00:00:00\n"
     ]
    }
   ],
   "source": [
    "print(pd.to_datetime('2016-03-01', format='%Y-%m-%d'))\n",
    "print(pd.to_datetime('01032016', format='%d%m%Y'))\n",
    "print(pd.to_datetime('01-mar-2016', format='%d-%b-%Y'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "* Задано несколько столбцов, скаждый из которых содержит элемент даты (столбец под день, месяц, год)\n",
    "\n",
    "    Тогда, `pd.to_datetime()` подается 3 столбца из таблицы, которые должны **обязательно** именоваться `['year', 'month', 'day']`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>year</th>\n",
       "      <th>month</th>\n",
       "      <th>day</th>\n",
       "      <th>day_of_week</th>\n",
       "      <th>births</th>\n",
       "      <th>date</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1994</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>8096</td>\n",
       "      <td>1994-01-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1994</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>7772</td>\n",
       "      <td>1994-01-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1994</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>10142</td>\n",
       "      <td>1994-01-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1994</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>11248</td>\n",
       "      <td>1994-01-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1994</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>11053</td>\n",
       "      <td>1994-01-05</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   year  month  day  day_of_week  births       date\n",
       "0  1994      1    1            6    8096 1994-01-01\n",
       "1  1994      1    2            7    7772 1994-01-02\n",
       "2  1994      1    3            1   10142 1994-01-03\n",
       "3  1994      1    4            2   11248 1994-01-04\n",
       "4  1994      1    5            3   11053 1994-01-05"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# переименовываем столбец\n",
    "df1 = df1.rename(columns={'date_of_month': 'day'})\n",
    "\n",
    "df1.loc[:, 'date'] = \\\n",
    "pd.to_datetime(df1.loc[:, ['year', 'month', 'day']])\n",
    "df1.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Довольно часто информация о дате передается в формате [unix-timestamp](http://www.onlineconversion.com/unix_time.htm).\n",
    "\n",
    "В этом случае можно воспользоваться функцией `pd.Timestamp.fromtimestamp()`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Timestamp('2016-01-26 03:00:00')"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.Timestamp.fromtimestamp(1453766400)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Давайте сделаем дату индексом в нашей таблице."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>year</th>\n",
       "      <th>month</th>\n",
       "      <th>day</th>\n",
       "      <th>day_of_week</th>\n",
       "      <th>births</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1994-01-01</th>\n",
       "      <td>1994</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>8096</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1994-01-02</th>\n",
       "      <td>1994</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>7772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1994-01-03</th>\n",
       "      <td>1994</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>10142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1994-01-04</th>\n",
       "      <td>1994</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>11248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1994-01-05</th>\n",
       "      <td>1994</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>11053</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            year  month  day  day_of_week  births\n",
       "date                                             \n",
       "1994-01-01  1994      1    1            6    8096\n",
       "1994-01-02  1994      1    2            7    7772\n",
       "1994-01-03  1994      1    3            1   10142\n",
       "1994-01-04  1994      1    4            2   11248\n",
       "1994-01-05  1994      1    5            3   11053"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1 = df1.set_index('date')\n",
    "df1.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Это дает нам возможность выбирать строчки по нужным датам..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "year           1994\n",
       "month             1\n",
       "day               1\n",
       "day_of_week       6\n",
       "births         8096\n",
       "Name: 1994-01-01 00:00:00, dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df1.loc['1994-01-01', :]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# df1.loc['1994-01-01':'1995-01-01', :]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ".. и не только, например"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Аггрегация с нужной частотой\n",
    "df_aggr = df1.resample(rule='AS')\\\n",
    "             .sum()"
   ]
  },
  {
   "cell_type": "raw",
   "metadata": {},
   "source": [
    "B       business day frequency\n",
    "C       custom business day frequency (experimental)\n",
    "D       calendar day frequency\n",
    "W       weekly frequency\n",
    "M       month end frequency\n",
    "SM      semi-month end frequency (15th and end of month)\n",
    "BM      business month end frequency\n",
    "CBM     custom business month end frequency\n",
    "MS      month start frequency\n",
    "SMS     semi-month start frequency (1st and 15th)\n",
    "BMS     business month start frequency\n",
    "CBMS    custom business month start frequency\n",
    "Q       quarter end frequency\n",
    "BQ      business quarter endfrequency\n",
    "QS      quarter start frequency\n",
    "BQS     business quarter start frequency\n",
    "A       year end frequency\n",
    "BA      business year end frequency\n",
    "AS      year start frequency\n",
    "BAS     business year start frequency\n",
    "BH      business hour frequency\n",
    "H       hourly frequency\n",
    "T       minutely frequency\n",
    "S       secondly frequency\n",
    "L       milliseonds\n",
    "U       microseconds\n",
    "N       nanoseconds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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>year</th>\n",
       "      <th>month</th>\n",
       "      <th>day</th>\n",
       "      <th>day_of_week</th>\n",
       "      <th>births</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1994-01-01</th>\n",
       "      <td>727810</td>\n",
       "      <td>2382</td>\n",
       "      <td>5738</td>\n",
       "      <td>1462</td>\n",
       "      <td>3952767</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1995-01-01</th>\n",
       "      <td>728175</td>\n",
       "      <td>2382</td>\n",
       "      <td>5738</td>\n",
       "      <td>1463</td>\n",
       "      <td>3899589</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1996-01-01</th>\n",
       "      <td>730536</td>\n",
       "      <td>2384</td>\n",
       "      <td>5767</td>\n",
       "      <td>1459</td>\n",
       "      <td>3891494</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1997-01-01</th>\n",
       "      <td>728905</td>\n",
       "      <td>2382</td>\n",
       "      <td>5738</td>\n",
       "      <td>1459</td>\n",
       "      <td>3880894</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1998-01-01</th>\n",
       "      <td>729270</td>\n",
       "      <td>2382</td>\n",
       "      <td>5738</td>\n",
       "      <td>1460</td>\n",
       "      <td>3941553</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              year  month   day  day_of_week   births\n",
       "date                                                 \n",
       "1994-01-01  727810   2382  5738         1462  3952767\n",
       "1995-01-01  728175   2382  5738         1463  3899589\n",
       "1996-01-01  730536   2384  5767         1459  3891494\n",
       "1997-01-01  728905   2382  5738         1459  3880894\n",
       "1998-01-01  729270   2382  5738         1460  3941553"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_aggr.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x11dc6c400>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Визуализация\n",
    "df1.births.plot(label=u'Исходный ряд')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x11e2c4198>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Визуализация\n",
    "df1.births.plot(label=u'Исходный ряд')\n",
    "df1.births.rolling(window=14).mean().plot(label=u'Rolling Mean')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Задание**<br/>\n",
    "Преобразуйте дату в таблице df2 аналогично df1. Становится ясно, что периоды данных в этих файлах пересекаются.\n",
    "\n",
    "Необходимо проверить, что показатели рождаемости слабо отличаются между двумя файлами.\n",
    "\n",
    "Сравните показатели рождаемости на датах на общих датах"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "# переименовываем столбец\n",
    "df2 = df2.rename(columns={'date_of_month': 'day'})\n",
    "\n",
    "df2.loc[:, 'date'] = \\\n",
    "pd.to_datetime(df2.loc[:, ['year', 'month', 'day']])\n",
    "\n",
    "df2 = df2.set_index('date')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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>year</th>\n",
       "      <th>month</th>\n",
       "      <th>day</th>\n",
       "      <th>day_of_week</th>\n",
       "      <th>births</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2000-01-01</th>\n",
       "      <td>2000</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>6</td>\n",
       "      <td>9083</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000-01-02</th>\n",
       "      <td>2000</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>7</td>\n",
       "      <td>8006</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000-01-03</th>\n",
       "      <td>2000</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>11363</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000-01-04</th>\n",
       "      <td>2000</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>13032</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2000-01-05</th>\n",
       "      <td>2000</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>12558</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            year  month  day  day_of_week  births\n",
       "date                                             \n",
       "2000-01-01  2000      1    1            6    9083\n",
       "2000-01-02  2000      1    2            7    8006\n",
       "2000-01-03  2000      1    3            1   11363\n",
       "2000-01-04  2000      1    4            2   13032\n",
       "2000-01-05  2000      1    5            3   12558"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df2.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_diff = \\\n",
    "df1.loc[:, ['births']].join(df2.loc[:, ['births']], how='inner', rsuffix='_2', lsuffix='_1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    1461.000000\n",
       "mean      223.484600\n",
       "std        68.774771\n",
       "min        60.000000\n",
       "25%       170.000000\n",
       "50%       231.000000\n",
       "75%       271.000000\n",
       "max       438.000000\n",
       "dtype: float64"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "abs(df_diff.births_1 - df_diff.births_2).describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Задание**<br/>\n",
    "Теперь надо аккуратно объединить таблицы, так чтобы соблюсти целостность информации (то есть чтобы на одну дату было по одному наблюдению - из одной из таблиц или среднее по обеим)\n",
    "\n",
    "**HINT**\n",
    "`join`, `concat`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Your code here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Задание**<br/>\n",
    "Найдите количество детей, рождающихся 6, 13 и 20 числа каждого месяца с учетом дня недели.\n",
    "\n",
    "Выделяется ли как-то пятница 13?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "## Your code here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Качество вина"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Загрузите [датасет](https://archive.ics.uci.edu/ml/machine-learning-databases/wine-quality/winequality-red.csv) с информацией о характеристиках вина и его качестве."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Your code here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Задание**<br/>\n",
    "* Что из себя представляет объект в этом наборе данных? Сколько их?\n",
    "* Какие признаки описывают объекты? Сколько их?\n",
    "* Какой признак является целевым?\n",
    "* Каковы их области значений?\n",
    "* Есть ли пропуски?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Your code and answers here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Задание**<br/>\n",
    "Какие признаки больше всего влияют на целевую переменную?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Your code here"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Вспоминаем NumPy \n",
    "## Линейная регрессия (для нас пока черный ящик)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "Загрузите [файл 1](https://www.dropbox.com/s/kg9px9v3xfysak9/tutorial_dataset.csv?dl=0) и [файл 2](https://www.dropbox.com/s/f87gm612o144emx/tutorial_dataset_2.csv?dl=0) в папку с тетрадкой. С помощью функции `loadtxt` в модуле `numpy` загрузите табличные данные одного из файлов. Присвойте y = D[:,0] а X = D[:, 1:].\n",
    "\n",
    "Сейчас мы воспользуемся одной магической формулой и построим модель линейной регрессии. Откуда эта формула берется мы узнаем на следующих занятиях.\n",
    "\n",
    "Модель линейной регрессии в матричном виде выглядит так: $\\hat{y} = X\\hat{\\beta}$, где\n",
    "\n",
    "$$ \\hat{\\beta} = (X^\\top X)^{-1} X^\\top y $$\n",
    "Остатки модели рассчитываются как\n",
    "$$ \\text{res} = y - \\hat{y} $$\n",
    "\n",
    "Итак, еще раз:\n",
    "\n",
    "**Задание**<br/>\n",
    "1. Загрузите данные\n",
    "2. Оцените веса $\\beta$ с помощью формулы\n",
    "3. Постройте график, на котором по оси Y: остатки, а по оси X: $\\hat{y}$"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "# load data\n",
    "D = np.loadtxt('./data/tutorial_dataset_2.csv', \n",
    "               skiprows=1, \n",
    "               delimiter=',')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = D[:, 0]\n",
    "X = D[:, 1:]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "Beta = np.linalg.inv(X.T.dot(X)).dot(X.T).dot(y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([6.85976194e-04, 1.17459428e+00, 1.08903184e+00, 8.87441294e-01,\n",
       "       1.00928516e+00])"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "Beta"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_hat = X.dot(Beta)\n",
    "res = y - y_hat"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x11edcb1d0>"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAs4AAAHVCAYAAADhFX3pAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi4zLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvIxREBQAAIABJREFUeJzsvXuQXVd97/k957S6ZUndrX6pW7JlsE1sY/CFGGMH44pJWeWihqorw6WAoOCoiCfFcC2EsCnANgoRNjiDfRVFDplkTBBmVBnymKCpykA5GgpPCRJbLoIjwJJtZGNbUkv9bj36ec6ZP3af1nnstfdaa6+191rnfD9VKrv347fW/q3f2nudvX+PXLlcLoMQQgghhBASST7rDhBCCCGEEOIDXDgTQgghhBAiARfOhBBCCCGESMCFMyGEEEIIIRJw4UwIIYQQQogEXDgTQgghhBAiARfOhBBCCCGESMCFMyGEEEIIIRJw4UwIIYQQQogEXDgTQgghhBAiQVvWHYjj5MmT2uf29/djdHTUYG9aC+pPH+ouGdSfPtRdMqg/fai7ZFB/+pjQ3YYNG6SO4xtnQgghhBBCJODCmRBCCCGEEAm4cCaEEEIIIUQCLpwJIYQQQgiRgAtnQgghhBBCJODCmRBCCCGEEAm4cCaEEEIIIUQCLpwJIYQQQgiRgAtnQgghhBBCJODCmRBCCCGEEAm4cCaEEEIIIUQCLpwJIYQQQgiRgAtnQgghhBBCJODCmRBCCCGEEAm4cCaEEEIIIUQCLpwJIYQQQgiRoC3rDrhMaWQYOLAf5clx5Nb2Apu3ID8w1DL7F4dPorRvrxX5NvvuwrlTg+tRev+HjbRrY1/abaqeg/5+J/rhwjk6ujPVRhbbAaQqo/7Yxa3bgLZ2ablJ2nJpW/nWO5A79FSi8yr3PR3Z9ceo/h0nEysvCfbPziRuL0qW7L6w46YOP43i6VPC45T+f3oy+NfVA3R16227ZDUwc17tv2engMUFoFwGCgWgVLr4/wsLwPwcUCoilPf9Tyhs+VT4PofIlcvlctadiOLkyZPa5/b392N0dFTr3NLIMMq7dwIjwxc3Dgwht2PX8k2x2ffn9/wpiqdPGJcPwFrfk8h28Vwb+9LuD+7aBjy5V+mcvl2PY3xiPPN+ZH2OjqzubV/G1N6vJm4jk+09/UAuB4yPpCMj5NjC4KUo/sGn5eQmbMupbflC7YLG5Hk6x6j+LSOzmqTtRcmS3adzXLOjuXhOst6rsGHDBqnjCl/5yle+kqgly5w9e1b73FWrVuHChQta55b/7q+BF39Zu/HCOeTOTSN3wy0tsb907IgV+Th2xFrfk8h28Vwb+9LuD178BTB6Wumc8tlJFH/5H5n3I+tzdGQt/PJnKFcv2DTbyGT77AVg5oLcsSZkhBxbPn9WXm7CtpzaVv8OzeR5Oseo/i0j02R7JvbpHNfsvPoS8v/195VPS7Leq9DZ2Sl1HF01BJQnxyO3c3+y/a7KdvFcG/vS7A8unFc+pzg+ivLCQub9yPocHVmlc+eMtJHZ9rRlJD3WxvmEEGfhwllAbm0vwn7/5db2cr+B/a7KdvFcG/vS7A9WLfm/KZxT6O3H4txc5v3I+hwdWfk1a1C6ELJ4Vmwjs+1h2JSR9Fgb5xNCnIVZNURs3hL4TFXT038xAGHzlsCXrZqBoabaXxi81I58m31vtnNt7Eu7P1u3K5+z+vf/2Il+ZH6OhqzOe75spo0stvf0A70D6ckIObYweKm83IRtObUtX7B3ns4xqn/LyDTZnol9OseRzGFwoIDSyDDKjz5QG2jQO4DcfQ+nkhnChf1rF+cxzqwaWueuHFyPOWbV0M4Mse6tb8fo6Gjm/XDhHB3dnXnhF85kyfAtq0bv1m2YZFYNrfMq9z0d2cyqcQdWHn4as6azanT3AJ3dett0s2oUF4DSUiaNcuni/y8uAHMRWTWWKPzv/3fk/jDSDA7kwllA6YnHUH7m6YbtuZtvQ/7ue7X75BMmDLFVoe6SQf3pQ90lg/rTh7pLRivpr/g//1fhPtcXznTVEKATPEUIIYQQQmLICZafou0O4X4PM6ISfCO7nRBCCCGENDdcOIvYvKUx0KB34KIfFSGEEEIIaSm4cI5CJQE6IYQQQghparhwFnFgPzBR52g+MRpsJ4QQQgghLQcXzgIYHEgIIYQQYoFcTm27Q3DhLIDBgYQQQgghFmjvCN/eIdjuEFw4i4irvEYIIYQQQtRZsSJ8e5tgu0Nw4SwgPzAE3LUN6FsXVMPpWwfcta2m8hohhBBCCFFkYUFtu0O0Zd0BVymNDANP7gXGzgQbZs4DT+5FaccuK4vn5fKsZ04FpS67eoCubmBuFjjxG6BYDD5trOkK+lLZDyyX7KwvqdwgW1B6WdSfqe8+juLpU5HnFI8eAfbtAS6cD/q3fiMwN1N7DUt9bCgF2tWD3LqhyP4syz83HWQ1GdxwMU2gpkwp2VXyQq8hRudZUTMeq1YDW7ejcO31ejIU9CKrC1nZE/0DKM7Pa42rDarnUIPNVcrMLi4E19TeAfT0W5+fsuf4Ooek7y0JbdIGqmOqcnzks+K148D8HNCxEth4RfBfnfutwv2juu9Tg+tRev+H5cezXAq1vyTjKG3vjt3Dq5+5oSW0x84AZ04FBULWdEqNjdNzaFGwQBZtdwiW3BaQZsnt0sgwyrt3AiPDyQQNDCFXt7APlR1yXGx/Qs4pHj0C7N4ZW3dep9+J5cdco+2+Z1E6NfSa8gVgxy7pxbPzOo9pwwa+zc9q23N+PAXt2J6fUSSdu6pjqnK8EVtUud/G3D9Se1ZIjqNWOxncU+rRGteYsclyDslQ3PYxYPZC445LVqHwF/+nsjyW3HaAVLNqHNif/KEMBDLq0+WFyQ47TuecfXuST8qo/iSRH3eNtvueBWHXVCoG25PIkCUNnWehb1/nJ+D+eIra8Xl+qo6pyvEmbFHFVuLuH2k9K2THUacdF+7hOuMaNzauzyH6ODcfaWbVMLkYr5el8wNA+pwL59U6F0FomwnlR+rVdt+zQHRNKtfqgc7T1re38xPwYjxD2/F4fqqOqcp2U9eiZCsRY5Hms0Lq2jXbyfoert1+1PW6PoeKgkW9aLtDcOEsIsWS2yYX4/WydH4ASJ+zarVa5yIIbTOh/Ei92u57FoiuSeVaPdB52vr2dn4CXoxnaDsez0/lMa34sEocb+palGwlYizSfFZIXbtmO1nfw7Xbj7pe1+eQKF9znnmc/Satktthqe90CEuXp/MDQPacrduBvAETEqX5SyK/YyXKt94h3m+771nwwU+EbMwF1yqL6zrPQt++zk/A/fEEwvXi8/xUGJ/SyHAQ0FePaDxN2KLq/Tb0vlLVnzSeFXG2WCGqr0llW6R86x1BIKcqA+sDGwrD9Tkk6lvO/WUps2qIiCq5bTg4MD8whNKOXbWR0t09QRTsyy/URpnm8sG+3n6gUzL6VecHgMQ5ub4BlNd0Bf2t0LYCGLosyPzR3VPTx4bo6e4e5AbEUd6h8gsF4LfeFtxk6iOOT74GlErB33OzkVlQYmVX9TH0GhyMyMY/fadxR2cXcn0DjdsFKOmlvQM4fiyI4geM6ry9bwDz1Vk1YmzFJjXzsz6rxvhokFGjJqAqD1z6pppI/izmJ+DvHFK6twjayXR+yo5p2HMGADZeEdpvpWdF2wrgLdddzKIgc7+9ZDVw/mztjv/rSZSuvEasxyTPisuvDB9HxXsLcDG7RwPrNwLr1gf/PzsTvOlUlG2T5Qxec7MXN65oB656a23WkbEzwPCJ2jE++jzKu3eG9t35OeSxqwYXzgLSLrmdHxhqWJCXnngM5frULOUScte8XT6zh84PANlzDuyvnZQAsLiA3KWXm8k8Eia/WESuu6dBfumJx1B+49XaYys30rC+KMj2AtED+OyU2o89VZ1XHj4VDOm8J4OsJFGEzU9AkH2nVEJug+QcsDk/K8f6OIds31tsojA+wufJ7IxQvPSzYnEBue61yN/9Vfl+1y+aAWB8JNoGkjwrBoZCx1P53lJpI+Tta+7yK2va0JJtk7B+L8wvjV3IONffb0R9d30O0VWj+XCh5LaJxbvN4CPbPy5MBM2Y2u46OuOpcix1Hk7Sa7IdHOjrePpsK7J9L40MA6OnQ49Vfc7YfFboyE/6rPD5uaWKK/M5dUTZMwrMquEvKQYHijCxeLcZfGT7x4WKfBOBN5FyXEdwnbH76qDO1Uh6TbaDA30dT59tRabvy3l7KwW2qtHwKbX5rACQ2AZU+6dqiyo/QlRlW0fg05t0jjo/hyouIbLbHYIL5yjSCg4UNX/rHUGS82ryBaVABh0ZsueY6F8USvI3bwmqtdXz+iuhwRO2+54mwgCjCgIdhEGdq5H0mmzOT2X5Do2nz7Yi1XdR3t6+dVrFJow9KyD4TJ7QBpT7p2CLyj9CFO3cJsWjR4Cj/9m4o7tXHBwq2Xfn59C5abXtDsGFs4go362UyB16KjQhfe7QU1ZlyJ5jon9RqMjPDwwFgSb1VPzzEsh2HpF/cwWBDsKgztVIek0256fqsS6Np8+2Etf30sgwyr/6efjJ/YNagVjGnhUQvBxKaAOq/VOyRcUfIap2bpV9eyDSuSg4VLbvzs+hej/zuO0OwYWzABf8g+jjrChfEFDjpf+XAjJ9tuZz26I6r9BMPs4AnBlPn20lqu/Lb0fPToUeo/sZ3baPs2i/1WeFpC0KZUT9CFGwc6uIipRELR4T6sWZOcTgwObDBf8g+jib8YtL6vvpOjJ9tuVz26o6r9BMPs62Zavgs61E9j2qtHKCfLnWfZyBRH60rj2HnLEvi0VnnLlGEQwObEIcCA400gebBRZs60hRvrI/Z9bja4qwa6mmp1/+uqhzNZJek+0CKL6Op8+2EtF34du+zm4t32aZNhPJqCbMB9jis8JqrE1YIZksiuWICrZEFHKRvl7X5xCDA5uUjIMDjfXBYoEF6zpSkK/s0+XC+Joiqu+iT2KysqjzaJJek835qXisU+Pps62I+i7KTnHdO5MXmbDxrKhG5ANs6VlhM9YmPzCE3I5dyN18G3DN9cjdfFuyHy6a5I48p7QdULxel+fQjMBNRbTdIbhwFuFAcKCRPujIkD3Hto4U5Sv5dLkwvqYwGBxInSuS9Jpszk8N+c6Mp8+2Iuh7+XtPqJXXNtBm4mdFHQ12YPFZYTvWJr9UfKVw38PI331vJlUDrcY4uD6HLhG4o4i2OwQrBwpwwbGewYFq8nNre0Pjk8N8ulwYX1NkGRzYqjqv0GzBga6Mp8+2Iuzj8WPhQYGC8tom2jQZHAig4Y25zWeFrC2q2Kxr6PRd9hzn5xDfODcfLjjWMzhQTb6Kr5sL42sMmcT9ksn9qXM1mi040JXx9NlWlPsYUV47aZtGgwOBBj9nq88KSR9d5/MVR2ExxsH5OeTxG2cjC+dvfvObuPvuu3HvveH1z8vlMv72b/8W27Ztw3333YfjxyOKNbiCA8EDLICiJl/F98vrm20VscVPKkgm96fO1WiqAihwZzx9tpXQviMnjDUwsZCx9qyop87ty/qzQsJH1/l8xXFYinFwfg61+hvn973vfbj//vuF+//jP/4Dw8PD+Iu/+Av88R//MZ544gkTzVolPzAE3LUN6FsX/ALqWwfctS1VPygWQFGTr/JpyvubbQUJv0QA0n7O1LkazVQABXBnPH22ldC+owxMTzYebOhljLVnRQjVtmD1WSHpo+u8S0IUFmMcnJ9Drf7G+brrrsOaNWuE+5977jn87u/+LnK5HK6++mqcP38eExMTJpq2RmlkGHhyb1DGc+Z88N8n96ZakpM+zoryRRHrTexva9qHkTpXo9l8nF0ZT59tRbqPmuW1Vdo07uMM1NiIzWeF7DnCN/ajp1Mvoa2Kz8/nxLT6G+c4xsfH0d9/sb56X18fxscdGTwRYYnqR4bTLblNH2fp7UKXBYG/mPP+X5KY9mGkztVoJh9nl8bTZ1uR7qNmeW2VNo37OAM1bl9OFCkJc6sEgLEzKO/e6fTi2efnc2I8fuOcSlaNcphfksDf6+DBgzh48CAA4JFHHqlZcKvS1tamff74+bNYCJN5/ix6E/RJhcWt2zDx66MojZ5e3pbvH0TP1m1ok+yDjgzZc0z0z1Tfp777OGZDXBbar7oGPW99eyLZOiSxPRXCrgNr+5Bva6vZllt5Cbo+8BGsjOmTKzpPS39JSWpHNuZnte5cGU9VbM/PKJLaXuicDGHl4Hp0G7oWW88KrGgHFuZrDxwfQccP/xHdO75i9VkhfU5/PxZ3PY6JP7kHpTONL7sqfXURn5/PSRmZn0EpZHt+fkZr/qX5zEhl4dzX14fR0Ys35LGxMfT09IQeu2nTJmzatGn57+rzVOnv79c+v7S6M3T74urORH1S6sPEOMrFWh+lUrGI8Ylx5NvarcmQPcdE/0z1vXj6VKiM+emp0PGy3fcktqdC2HUgl0Ppzj8AvvuXwNwsAKA8O4OpvV/FdMynYVd0npb+kpLUjmzMz2rduTKeqtien1Ektb3QOVlPx0rMvf/Dxmzc1rMCpbClDTB7+hQWRketPiuUzmlrR6lnAKhfOFf11UV8fj4npdQe7hZWar9Ea16YeGZs2LBB6rhUFs433ngjfvjDH+K9730vXnrpJaxatUq4cHaGzVuAl34VBFVVSLtcZVQQwN3hGUxkZZQf+hyKl6wCunqArqUSl9OTwb/ZGeD82fh2RbK//nkUr7wm+Ht25qJP3OxM8Jlo8xa5T5Qi+X/2RRTXXwac+A1QLALtHcsLxHqEn6VU9KLa7zQR2cg/f7dRJxVXoyjbidP5a8eB+TmgYyVQCI/AN6Hz8eJi8OPVRZ1Xk3SOysyhyrys6Oc3vwYmx+TazGIOCfqeWzeUfO6burfYRCZgd8PlZvtp61lRXAw/tqJ32XZ1+qd4jii/Md54FcXHHwr+f3YmyG4yfCKw91Wrga3bUbj2+vA+2MamXlyfQx77OBtZOP/5n/85fvWrX+Hs2bP41Kc+hY985CNYXAwm3B133IHf/u3fxs9+9jN85jOfQXt7Oz796U+baNY+GZerNBLwEfILHABw4Vzwb+yMvKw6XzGh7LNTwPPPhssAgOPHUJIIihHKnxoP/lW4cC78uIgfOqp6Uel3mght4UL4zSfOdqR1vhjmyARjOl+W7qDOq0kcHKgyh2Lmav38jJRvcw4J+l5+5VjyuW/o3mITmbHPrVufSptGggMLheDHVTVLfs4uBAcus3kL8OIvGxeL588KbQYz54HdO1HcsSuTxbPV4EDX51Cr+zh/9rOfjdyfy+Vw9913m2gqPUz8gk+IkYpI0wazl0zVydKVLfPmM4n8ClEVuXRky/Y7TUSFTVatDv3lHms71LkSieeozflpQn5W42n73mIR4ZvPCh0rjX+5NPGsEPa7c23jF46l9JY2q/upnpMfGELx8ivl0nNWUyoC+/YAj6SfJteqXlyfQx6/cWblQAEupHIxksDc5K+3elkJZEvpMWnfoypyacp2JpUPYrIgfPATerZDnSuReI7anJ8m5Gc1nrbvLTbZvCVwfRFx6ZvMv83TqUBXh8iW0dkVfvzkuN3qfjrXpFuFUfCFzjoWKwc6P4c8fuPMhbMAF1K5GElgbvLXW72sBLKl9Jiw75FtaMp2JpUPIPal3HgFckee07Md6lyJxHPU5vw0ID+z8bR9b7GNyDcYADq77bSZ0LVQZMvCRWXla5fN6n6K16Q99lE/dGxjqXKg83OIb5ybEAO/4JNi5K1311pDvQmRpStbtlpWkr7HtaEjO+WS63EI7WB2Rt92qHMlEs9Rm/Mzqfwsx9P2vcUi5e890egTXE1VHmRj6FSgq0Nos11rgZ6QNF+vvxJcq63qfjrXJMrpLEEm+Z4tVg50fg55/MY5lawa3pJxcKARv7V161F+5cXGHavWAJesArp7Lr4BqUTAz80A5842nFIf0CKU3dkNGIjaFcrv7gXKpfAStqs7kXv7DbFtKOnFpYj9KuLsQ8d2tHTe2Y3cde80qvMVxUUsepBVI+kclZpDlXlZ0U9YVg2EB5xlMocEfc8NyGfVsH1vscrxY9H7l/yDTfqQmnjJIrTldetR7lrbuFgbH2nM8SxoV2ee6FxTfmAIpR27gAP7UT5zqnbeAIHNnHq90e6nxjPxjbdaUdH1OeTxG2cunEU4EBxoJCWeQEbuvoeFk6M0Mozyow/Et6shW4kI+eXv7A1/6F/2ZuRlxsd239Mgzj6OH6utftmxMt73VkfnGy43rvNeT/I4J56jNudnjHyn51AzzM8Iyr/6OUojw0YrByYOJI+y5e/sVe6PtOwIGTrXlB8YinxGFx99INTus/CNtxoc6Poc6lobninI5Fc4S9BVQ4ALwYFBgwbeetvyodKVrYJAvhEf9Iy/KBihvs+LCyh/74lgUdS3rtZ3b24WeHJv/CdJ6lyNpNdkc35GHOv8ePpqK5e+Of6Ys1NGy0EbCSQHlG0Fl75Jvl3F8TR2TXUIr0WUocgiOteodI7Lc0i0QObC2V9cCA404bdm1YfKRP+iiJIf5ssm80ZVRrYvhF3D9GSQo/PYEeDofwbFSqqppBpSkUmdi0l6TTbnZ9yxLo+nz7bSIRloFjcXFTASSK5hK0BOrl2N8TRyTWFs3iL02U7bz1nnGqXP8XkOOQ4XziKaJDjQaoJ1y2/lo+TnB4aAu7Yt3byXkH2jGiPbF8pnwkskx56nOfbUeSOJC6BYLgzh63h6bSsKKdHSuFeakCG0lZd/JSUrlQIokuQHhoDLr2zcUfE9TxGfn8+JmZ4K335WsN0huHCOwoHgQJXtpmTInmP7rXyc/Nyhp8RlpRPK9oIw/1QJkox9y+u8jqTXZHN+yhzr6nj6bCtKfTTkHpDGsyLUVgQVRE08K6zagODHTdqLSp+fz4kRFWgJK+TkGFw4i3DhM4eJt942E6zbfisfIz/RL2oHvigkZpVG2p5CIdHYt7zO60l6TTbnp8Sxzo6nz7YS1veunuBfPYbcA4z4A2vaCtrqcgyYelZYtAFnFpU+P5+T4nE6Oi6cBTjzmYPBgcK/Ewd5uBw4EUNpZBjQcdUoSCTSoc7V8DQ4EHB8PH22lfq+FgrAhssbjzPkHmDMH1jHVhbrir2YfFbYsgGXFpU+P5+T4HE6Oi6cBTjxi5TBgdHykwR5uPBFIQkH9jd+NgWAFe3AO24CrrgayIdM7/m5ZGPfyjoPw+fgQMDd8fTZVkR9P/Fq6OGV1HRJMPKiR8ZWZNKYmXpW2LYBFxaVPj+fk8KsGk2IA79IGRwYLT9JkIczXxQ0EfbzsjejcM+DKNz/KPCm31I7N2IfdR6Oz8GBgLvj6bOtKPfRQGo6Ey96ZGwlt2NXkOZSUZZLwYEAnFlU+vx8TgwXzk0KgwMjt2cdHAhAO8jDiS8KSRB8Sq+uHpdbJ3g7FPEZnjpXw/fgQABOjqfPtiLs45XXiN/YJk1NZ+BFj4zO8wNDQP+gsizXggNdWVT6/HxuZbhwFuHAL1ITAR82E6zbSlCvIl/XR9N2321SGhkGXjveuKP+QanxGZ46VyPpNdkugODrePpsK6K+Y9Pm4I1tpfxz/XlJF20JX/TI6jx24WXqWbF5S1Aavpru3tAfA6WRYZSeeAzFRx9A6YnHYt/eO7Oo1A0OlNCL83OI6eiaDxd+kZoI+LCZYN1agnoV+Zo+mrb7bpWwH3UAsPGKmlKqOp/hqXM1kl6T1QIIssc6OJ4+20pU3/MDQ8hd987wE5OkpjPwokda5yJ7iThHZzzLYyONqcmmJoLt1WJGhlHevRPlZ54Gjh1B+ZmnY11fnFpUqlZUlNSL83OI6eiaDxd+kdLHOV6+ro+mCz+MdBH2MeyTu+JneOpcDd99nAE3x9NnW4ntu4XKdWk+K4T2EnGOVv/27QFQv4gsL22v4sD+wNWlmhjXF2cWlTo/eCT14vwcYjq6JsSB4ED6OEvK1/DRdOGHkS7GfVx1treYzkU0hY8z4Nx4+mwrcX23Ubku9WdFTHVEI8+KC4K0ZHXbnQs8VECrH5J6cX4OMR1dk5J1uhoWQJGSr+Wj6cAPI21UC2DUv93q6RdfJ3WuhucFUCo4N54+24pM30U/VHRT06X8rIhcfJl6VlSX966mrvCTsC+jp8W6FAVXp7yo1Jp3knpxfg4xq0YT4kBwIAAzi3efE6zLyNf99Jn1D6MkqPQ9l4v+W0d2K+pchMcFUJZxcTx9tpWYvgsXTElS06X5rIjyc64viKIqG0sB0PWuFECQm37r9sa+hGUrGTsTqkvp4Oo0UJx3SnoB3J5DXDg3H058ymEBFCn5Wp8+XflhpINqAYzx2qARE3ppOZ2L8L0AyhLOjafPtiLT96hiIjqp6VJ+VkT6OU9PJH9WHNgPTE82br/67Shce31DX4T5pcN0KRlcnQbK805BL17PIcfhwlmAC/5BDA5UkG8hCM5VTAeHaW9vIZ2LaIbgwGUcGk+fbUU2INNkarpMnhURfs5JnxXCNgVvTKPyS8v2Jc5v2xoK805FL87PIY/T0bVl3QFn2bwFeOlXtW/rMggODLtNqAZ8qMqQPUd0HF5/BcUv3g109QBdSw+G2ZmLflvTk8G/rp6gSMfmLaG/9FX6LuxLhC9b6PHDJ1B8/KGaPoZew+xM0A9B362i4J9nSi9p63yifwDF+XlpW8mKpHPU5vw0dWwWc0jp3iJoJ6v5Kavz/MAQSte9M0ijVk+Ej2tp6S1q+cypi9cd9gYVAE6fRPFr9xm/30YdH3aO8NiTrwW2AtSOo+B6dOYFXjt+sY3ZGeD0SWXZNhH2+9TrKD76QO1z88RvhDJk5ToTHOhxOjounKPI2D+ofOsdwOFDtT5NOgVQFGXInhN6HABcOBf8GzsT3bmxMyi/cgw4fgylHbsabuZKfd+8BXjxl4033CVfMSnZADA1Djz/bE0fRZQBYd9toeyfZ0IvGeh8/pVjtcfE2EpWJJ2jNuensnyH5pDWvaVuWxbzE7Cr80rO4pr0a1H32cmx4N/Sccbut5W+v/B8o+tAPi//rDg7VWsrUdcjEzQb1p+Z841tqMq2iajflR8ScQwMhfbdxPrBKl1rw8eaPs4D3/QcAAAgAElEQVQe44B/kJcFUHQQ+PSp9F3VV8x2362h6J9nRC8u6TxtfcfQFAVQlnBpDnk7P2FX56E5i1UxcL8Flvp+xdWNO0olO8+KGB/k/MBQY1ETQ7JtkqjffeuQE/wwdCZXtQgGBzYfLvgH+erjrIMJ/1wjvmIauGATkf55BvTiks6d8dFDk/k4A87MIV/nZ1R7LunciK0A0n030m8ZH+S5WXuybaLb7/5B4YLfhTVMs8KFswAXggN9LYCiQ6rFOyKO1cF1m3CiYErEsao446OHJiqAYlC2DmnfW2zig85N2IrKOTaL4tRQn8vYpGybWOi3C2uYSEQ/VrL+ESMBF84iXEgeHpaySODPZFSG7DlRKZVUEPVHte8qx9vuuy1sjqeO/LR1nra+40g6R1t9PEXyfZ2fgPs6N2UrKuck7bfsOG7dru724MI9xUa/TawfbCIKgI0q/uIIuXLZpYzYjZw8GR4BK0N/fz9GRwXRxjGURoZR/sb9tf6kPf3Iff5rqfpCLUdQT45rR4nryCiNDKPjh/+I2dOnIs+plt2QNaO752K6pbCsGt09yA1EZ0pQ7bvK8Q3R6dX9jbuGmKj9JLYXh+54KuslQ5239w1gvjqrhoStZEXSOWp6POttz4Xx1JlD0vcWQTu690sTczcTnc/NBoHD83NBdbmNVwT/tXi/rT9n5eB6zL3/w/LjKXguJBnH4tEjwL49wLnpIKB/cMPFl2CuZEYKoXj0CPLf3YvS2bNAewewfiNQLmk9NyuYWD/Yovj4Q+FBm++4CYV7HlSWZ2LebtiwQeo4LpwFlJ54LDRNUO7m25C/+17tPvmEzcVfs0PdJYP604e6Swb1pw91l4xW0l/xa58H6rMnAcCV16DwpW8oy0tz4UxXDQF0rCeEEEIIsYDHeZy5cBbgvGM9IYQQQoiPMB1dE+JCcCAhhBBCSLPBhXOTknHlQEIIIYQQ4g5cOItwoHIgIYQQQkjTwTzOzQeDAwkhhBBCLOBxHmcunAUwOJAQQgghhFTDhbMIBgcSQgghhJiHrhpNCoMDCSGEEELMQleNJoTBgYQQQgghpAounAUwOJAQQgghxALTU+Hbzwq2OwQXzgIYHEgIIYQQYgGW3G5CGBxICCGEEGIeVg5sUhgcSAghhBBiFi6cmxAGBxJCCCGEkCq4cBbA4EBCCCGEEAswj3PzweBAQgghhBALMI9zE8LgQEIIIYQQUgUXzlEwOJAQQgghxCx01WhCGBxICCGEEGIej1012rLugKukHRxYGhkGDuxHeXI88KPevAX5gSHjspeNcnoy+HfJ6qBSz+JC8Ea9vQPo6Qe6ujHR3oHi9GRkf4pHjwD79gAXzgfnrt8IzM0Esrt6gK7u4MDZmca2u3qQWzckda0q7ZjWn2vYtJUa+WdO1ep3bhZ47TgwPwd0rAQ2XhH8V2E8l8fx3HRgb4MbLrpEVcmZ6B9AcX7enK2USxftr9oWW8BeInWecF7GtnnhPLBqNbB1OwrXXm9dhuw5OrJtzztZMnlWjI/WPiPWdAEz5xvuvVOD61F6/4eTPSuq7FDnvi5t7y0w94kdcuWy2/4HJ0+e1D63v78fo6Oj8QeGUHriMZSfebphe+7m25C/+17tPoW2NTKM8u6dwMjwxY0DQ8jt2JV4QofK1iGkP8WjR4DdO4FS0bjsarTaMaQ/XZLYXhQ2bUUoXxVBf4zYiw1bkZDrE9W2l0jnmjoJbTNfAHbskl4868iQPSfuuLC5a3veycJnRfM9K6qx9dxwkeKjDwDHjjTuuOZ6FO57WFmeCd1t2LBB6ji6aggo33pHcDOtJl8ItpvmwP7Gm9XSL38rsnUI68++PclvhCLZSdsxpT/XsGkrIvmqiPpjwl5s2IqMXF9JonNdnYS1WSoG223KkD1HR7bteScLnxV8VjQLHrtqcOEsIHfoqdCba+7QU8bbsukWYtK1pEHWhfP2ZBtopxlzbtt2IbIqx5C92LCVWLm+klDnWjoRtanSFx0ZsudoyHYlrz+fFXxWkOzhwllAmjdKmzmjTeadbpC1arU92dV0rNQT6sEvV1Vs5xe3KseQvVixFaAp7SWxznV0ImqzvcOuDNHY18vSkO1KXn8+K/isaBqYVaP5SPNGadUtZPMWwITv1sBQYw7rD34iuVyR7CVKI8P6n/hefyU4v4mw7kJkwl5E42nCXmzZCtCU9pJY5zo6EbVZKsrLUpRRGhkGFublZGn0L1XXvQj4rOCzomnw2FXDSFaNn//85/j2t7+NUqmE22+/HXfeeWfN/h//+Mf47ne/i97eYNH5/ve/H7fffruJpu2xeQvw0q+A8ZGL2ywVQMkdegplkVuIYiR6PfmBIZR27AqPlD7xmyBDQj2rO4G3vBXt7R2Yj8iqkTvyHEIjS1etAS5ZBXT3AJ0RWTW6e5AbiIneP7A/OLaerrVA37plORgfBSbHao8ZHwnONxzMmSU2bQWos5dKVg2RfgFgbR/Q0yc1nkJ76e4F3vyWGrto7xvAfHVWjaS2sn5jbVaN0ydbw17idF49L19+ATh/tvY4DZ0I2zw7JS1LWcaB/Y19r5KFm2+Tl/3WrzfKsDzvZMnkWRFmF0Dwhv6yN9fc41cOrsdcSFYNpWdF1Zyvlh2bAUM0/y9ZDVz9tmU5rTL3iT0SL5xLpRK+9a1v4cEHH0RfXx++9KUv4cYbb8Rll11Wc9wtt9yCP/qjP0raXLqkVADFtltIfmAo9IYgjGq97M0o3PMgemKiVIX923iFVlSsUhvrN9a0UXz0gdCFXbP5rqXhQhRmLyL9YnCD9FgL+zh0KQr3PFizKc72lOTX2QpAewnTueh+oKqTqONlZanKUDlep3+t4OMMRMz9sOfEFVc3zKtuwbzN9Flx+ZU1tt4qc995WtlV4+WXX8bQ0BAGBwfR1taGW265BYcPHzbRt2xJsQBKVv5zSdtNpd+Czzb1bQjb9OCzjwq+2oopGZFI2kpkmy1sL6Z0EjWeSe8ton0qx0f2QcGGYmVZIIt++DD3SyPDwOhpqTZaZe47Tyu7aoyPj6Ovr2/5776+Prz00ksNxz3zzDN44YUXsH79evzhH/4h+vv7Q+UdPHgQBw8eBAA88sgjwuNkaGtr0z5//PxZLITJPH8WvQn6FMbi1m2Y+PVRlKomfr5/ED1bt6HNcFsq7cbpz3a/F4dPYvyNVxs+8YW1sbh1G8ZffgHlsTO1x574DdYuzqNtSC4/oymS2F4UrtqKaRmq+lOxlUpfXLIXk1TrTkXnpnQy+4GPYOo//r3RDWxtH3ol7WVx6zaMHftF45tBgQzR8bne/objhbIRXGt59DT6+wcb5Gcx7+rJoh8m5q3Nfi8On8Tknj9Fsc5uAaAweCnWOv6sqMbWc8NFJtrbERaV0N7ejh4NHaSpu8QL57D6Kblcrubvd73rXXjve9+LFStW4KmnnsJf/uVf4k/+5E9C5W3atAmbNm1a/jtJQutEBVBWd4ZuX1zdaTxBeWliHOVird9aqVjE+MQ48m3tRttSaTdOf7b7Xdq3t+HmBgClS9+EybZ2oLpvbe0oX/ZmoO740uhpjO/ba7xoTRzWCqA4aiumZajqT8lWAOfsxSTVulMaNwM6KY0Mo7z3q+GxE/m8tL2UJsbDA73K5VAZouPLi43XWpoYB+qeUcv7Rk9j+rt/hYVP3NMgP4t519C/DPphYt7a7Hdp316UT59o3NG3DqXtf+L8s6KaliqAMj0Vun1+ekpLB14VQOnr68PY2MVf7mNjY+jp6ak5prOzEytWrAAQLIyPHz+etFn7bN5ysUxnBUvBgWm6hRht13K/hT5nir5RTeW75qutmJIhQNlWIva1tL0k1UlUEY1KAJasnLBAr6nxcBmi46cnGo8P00kVxfGQfVnNOxf64evc7x8UBxK2wtx3HY9dNRIvnK+66iqcOnUKZ86cweLiIn7605/ixhtvrDlmYmJi+f+fe+65hsBBZ2mS4EBb7drut6pfnCt+iDbx1VZMyRChM/a0l0aS6iRuLF0PDgSA4pmTDanJWiU40FabnPukmUi8cC4UCvjkJz+Jhx9+GDt27MB73vMebNy4Ed/73vfw3HPPAQB+8IMf4HOf+xw+//nP4wc/+AE+/elPJ+64dRgcaP38OJRzloblIY3I++kjvtqKKRkitPLb0l4aSfqlLe5tkeTbJOUAvgi5SsGBAEpnhlHevbNm8ezKQovBgY1w7nuKx1k1jORxvuGGG3DDDTfUbPvoRz+6/P8f//jH8fGPf9xEU6mR5i/78q13AIcP1fropZBcP2m7tvutmrO0Pg9pbN5PD/HVVkzJEKGT35b2IjpJ70tbaWQYeC3GDW+p0EScjsu33gE88/8BYdl/62REtpvPN1xrqE7qGRmuyeub1byrJ4t+cO4TK7Syq0azkuYv+9yhpxpv4pWJb5Gk7drut86Pl/zAEPJ334vCfQ8jf/e9TXcj9NVWTMkQoftDl/ZSR5IvbTG+wwCk/ZyD/gkW7PUyototlRquNVQnIVTbTlbzrp4s+sG5T0gtRt44NyUpVg701W81DR/nsEen7c+jxaNHgCceA85NAbk88JbrkLvrvztxY/XVVkzJEJGVrbiOqYIeMmNkwn9Z9pjq/ap+1dL2VvXmiz7Oydrk3CcNeOyqwTfOUaQUHOir36pzPs4GKB49Ajz2YBC9XywCiwvA0edR/rMvNgQMZYGvtmJKhogsbMUH0gywNVHcRPaY6v0qx8q2D2DZJSTqHPo4pydDBOe+n7gyp3TgwllEmml/0kx9Z7Jdy/3O5PPovj0I/UQsSoOVNr7aiikZAlz5lO4cijpPtAgJaytf94jpWKkvq0JPf03/y7feAXSsDD827FrDZIflEq52Cclq3tWTRT8494kFfP7Bw4WzgNQ/iaX0dtt4uxb7ncnn0QvnhbucyfHpq62YkhEm1pFP6U6ioPPEi5B62ZesBto7Lv49Nws8uVfu642on1XFS0ojw8CTewO5KjIadFIKP6zafrKad/Vk0Q/OfWIYn3/wcOEsINXPCL4WtbDcb6GubUbdrlot3OXEJyRfbcWUDAGZ2IoPKOo80SIkrK3zZxurCFYyVqjKqlD9Jjiq4AoQfq1hsouL4edX7IcFUJK1mcXcHz3thHsdCcfnHzxcOAtI8zOCrwFf1vu9eUvwWbaeKt9D42zdDiCkHG93rxM5Pn21FVMyhGRhKx6QZgEUE7agul8n0FB4TqHQuG3Jflx5yDM4MISwnMwAMHamIR83cQimo2s+0vyM4GvAl+1+5weGgMuvbNyhUrpXkcK11wP3PhQslAsFoG0FcO07kPvCI05k1fDVVkzJEJGFrfhAqgVQFB54qgF9orZ0Ag2F53Subdy2ZD+uBDIxOLCR/MAQcjt2AX3rGnfKfN0gRBEunAWkXgAlAyf5pO2m0m9Bahqbb1gK116PwqP7UPjf/hmFv/onFO79qhOLZsBfWzElI5IMbMV1tHSu4YsqLEKyuqsxQFBizEP7XU3lTXDccSFtiXSCzq7wvkyOOxPIlEU/fJj7+YEhoH8wvO0Wnv9O43E6OuZxFpBmbkidykcutCs6H3/zv6J45TXB37MzF99Ezc4oV2gSjQNOn0Txa/cB05NAVw/Q0QEMnwiChFatBrZuD94eNxm+2kqUDPzV11G8ZFUwjl3dAIDx4iJKqzvN2cqjD9TYIXI52kvY9Ub5oi5V0QtF5JO8YgVwvi7wTsJuQvtdTeVNMBB9XEhbQjsUBQavvCSzeVdPFv2wOvfTeFa88SqKjz90sY0WmfvO47GrBhfOIlgAJf78MwLfsbNTwPPPhp8DAMePobRjl9wNcfMW4MVfNj6UJ8eCfwAwdqZ238x5YPdOFHfsaroboq+2AkTYy4Vzwb+qcVyo/I9pWwmD9qK9PXa/INtFUh9n2WPCjhOe17UWKJUa7ef1V1DuDn9hQh9nSRlZPivOnxW20cxzn9iDrhpRsABKNNMTeg0r+J0JfVfjKBWXcjI3F97aCqBnL2nYCkB70dweu1+QpSaxj/PSMSZ9nHPr1ov95AW2Sx9nSfisIPV47KrBhbMIFkCJpyskmEYSpTckuhMpIiezt/hqK4C2vaRiKwDtRef4uPO2bm/MeCBTBCWqAEqVjMjiJ6K+R12jyH661rIASpI2+awg9XjsqsGFswAWQJEgwc3QRER2LNWFF5oJH20F0LaXVGwFoL3oHh9xXq5vALhrW+3iVrYISlS7c7PA3+4Gvv3n4uInUTIE1yi0n46VLICSpE0+K0gTwYWzABZAscjAkHq5Vs2sFk2Xw5O2Ek0CWwFoL9rjHHFe7tBTjYvbuE/wUQVQquVXx6CIjpEpgFI5bvMW5MLSmh0/xgIoabZZgc+K5sVjVw0GB4pgcGA8IgO/ZDVw9dsuHpMgUhoIfNdKO3YBB/YHfavIm54M/nX3AKOng/+vZmo8PhuAZ3hrK4DYXjpWAmu6gnHsDLJqrCguYlExqwYQYSt1dohTr9NeDGxPcp7uPlVkgwPLk+MoDAyhcOU1mK8POK6vfhgjyxa+Bgdm8qw4c+ri82HpvtJKc9910sxcZhounKNIMTgwCwNK2q7w/P9yI/KGb0D5gaHIm1rx0Qcab4ZovhyevtpKpIx33txgL739/RgdjXnjKCDOVgDaS9QXNa1xFvglVs5TlSlMLaZBWHBgVH/KM/L+rlkEB6Y9/63OfT4rWpbyrXcAhw/VFprLIDe6DnTVEJHi5ylfi1q4UhQAyC7bRNr4aiumZJiC9hKuc50xEhY/WfpCpyMztrBJhVzMI0yhAErluEJvSOl2Sdm2YQGU5LTK3HedNKszm4YLZwFpfhLLyoCStuuU4Yf5tslE74dQPHoExfu2ovipD6L4v/w3FB/7sjP+b77aiikZxjBoLy6jqnOtMRL5Iy8uoPydvcC3/oeyzNB+hFEuRe8PaSfuGlf//h832kZ99UOBbNtkMYc494kNsnI7NAEXzgLS/FXqq9+qS4afHxjSj96vonj0CPDYg4HPW7EILC4AR59H+c++6MTi2VdbMSXDFKbsxXXS8HEW7pueBI4dERafcdHHGQDahjY02kYpfIFOH+f0ZJiiVea+8zAdXfOR5qclX4tauPbJSyt6v559exDqkVkJHskYX23FlAyTGLEXx0mjAIru+NmQKSNL5hpDbUNCtm18LYDCuU+aCS6cBaT6acnXohZZ9VuAkbcaEYnwnfiE5KutmJJhEJfeglkjjQIom7cAbYpx5rlc9EuIsH7Uu0sU2oAV7dHHqBZAWULKBlgAJV0ZBmmJue86Hqej48JZAAugpHS+QYy81RCUCFaWYxNfbcWUDEO49hbMGmkUQJEJ5KuXGfdpvL7dS1bXFqooLgIL8xf/XtEOrFoTLUO0ve5vKRtgAZT0ZRiiZea+y9BVo/lgAZQUzjeMEfeardsB5Bq3d/dm9nakBl9txZQMg7gU6W+NNAqgHNgvzHMcSdSn8bB+nD8b3c7CPHBuunabagGUCpu3AD0x2TVYACVdGQZpiblPrMGFs4gUPy35GvDl2ucuE+41hWuvB+59KFgoFwpA2wrg2ncg94VHlBLx28JXWzElwyRORfpbItPgwNWdwDXXA++4Kfh/Q+2pohocCCwFkF1+pbJs2zA40AytMPedx2NXDRZAiYIFUKyebxpTN+fCtdcDj+4z0CPz+GorpmSYxLWHuQ1SKYAiKn7y9huWi1uUnngM5WeeTtwPHVQLoCwj8QBnAZT0ZJikFea+67hmEyrwjbMIFkCxfr4MpZFhlJ54DMVHH0DpiccifSJbwW/NV1sxJSMKFVsBaC+hbN4SfG2pJsJNKa74iW4/hAVQ4gqe1O/XKICyLCrODlgAJVUZUXDu+4fP7jJcOAtgART758dRGhlGeffO4E3VsSMoP/M0yrt3Cm+KPk9EWXy1FVMyRKjaCkB7CaM8NgJMTdRunJoItochKn6y8Yoa1yYjhVgAxBY8qd+vUQBlmTg/ZxZASVWGCM59P/HZXYYLZwEsgGL//FgO7A8CiKqJCCjyeSLK4qutmJIhRNFWANpLKKF5zMtL2xXk1Lk5mPSpVkXHxxmQ83Omj3N6MoRw7nuJz+4yXDgLYAEU++fHkUZgk2/4aiumZIgwGdTW0vYiymMu2C4r32TBFVUS9SXGz5kFUNKTIYJz31OYjq75YAGUFM6PQXhTVZxwTeW35qutmJIhQNlWIvbp2Iuqj2VqqOq8ugxxNYL85uVb72g8Z2AovOhIfVaajpXRvtb1/e7qadxWTU9/o3+2ZgGUCpG2wAIo6coQIByj0dPiedgKc59YgwtnASyAktL5UYh8DF9/peHmJBuk1BT4aiumZIShYCuAWXvR8bFMFUmdl0aGw/2K8/ml/OYhxz+5t7Z0ccdK4K5tDakb8wNDwF3bahfZc7PRRVDq+1koAB+6q7FaIBAURvlvf9hYOVCzAMoyUX7Oi4vh223DAii1hP0oA4CxM6HzsKXmvst4nI6OC2cBLICSwvkxCH0Mx0fkihoADUFK3uOrrZiSIUDJVkR9AfTsRcPHMjVUdH5gPzA92bj96rcHKRrDjq+/7rnZ6IC/6kU2INaTqN///N3aaoEV5ueCfTLXqqCTSD/n6QkWQElThoD8wBByO3YBfesad4bZV6vMfcfxObMJF84iWADF+vlSCH59ygb8+PDrVQVfbcWUjEgkbSWyTQ17cdlfUqVvwv4K3gzajEEQ9kXkgx2xTzc4cJkIm2BwYHoyosgPDAH9g1JttMrcdx2fM5uwAEoUGRdAwfAJFB9/KJjQFZ+s6cngX1cP0NUdbJudCX6lbd4S+ou5tPQLuDw5XivnxG/CO/TGqyg+/hAm2jtQnJ4Uyhb2+/VXUPzi3Q19DLuG3LohYb8j2zj1Oopfu++iLsLeIMCPX68qKOlcw1aAKns5cypWvzh9smYcosYz1s6r+jvRP4Di/Lw5W3n0gYv2NzsDnD4plKGMw771SveWN14VylCRrVzUJER/wmNXrQZmBItnwT7tAihxfYk4xxYqcyjsPhA3h0KfFQK7wCsvovi1z9fcW6YG16P0/g8ne1Zo3reEbbx2PNDNkpxWmfuukzv0FMqiOLKwL1wOwYWziKhPS0vVsExRvvUO4PChRv/CqXHg+WfDTxo7UysDAI4fQ2nHrpobS8UHq+FzUhTnzwLPP4vKB1GRbGG/L5wL/tX1Mewayq8cC5W9zOYtwAvPN35Crtxcl+SE0oT+zVo6l7QVQGAvUeM4ORb8Wzouajyl7XzsDOZfOdZwDUZsJQpNH0eXfeu17i3VRFxHqOy44iov/rLxvrrkhx57b8kXgJvfB/w/f98ou7Mb+OAngL/989j+aPU7zK7y+WwKoEjOoQZi5pDys2J+Dqibp7PHjgAv/KeZZ4XCfQuAeJxmzsfbehPOfdfx+W09XTUEZF4ARQeRP5eJQIUQ2Vb7vUR+YCi8gpgMzebfjBR0bsJeBLKN9N2WrQD6Po4O+9Yn1nnEdahmHlLxQxfJxsED4f3MF5A78pxUf7T6fcXVjTtKJTcKoKhic+4L5Dfls8Lxue88TEfXfLhQAMWELF9l11AfUCRLk/k3A/Z1bkp+2rKX0bUVwKiPoyu2l1jnGv69kW0mjVlYWAjfPj8n3R+b/baND/MzTFYzPitcn/vEHlw4C3ChAIoJWb7KrkGQQzaRTE+xrXNT8tOWvYymrcTKDaE0MgyMnjYiyxZJ+xF1vs7LhaQFU7BiRfj2VautFWPRPccGPszPMFlN+aygf3MymI6u+Ui9AIqJTzuyhQd8kF3N1u3qn+DiZPqKbZ2bkJ+F7Ao6tiIjt45lf9Aw30yXbC+JzuOuI0y2qXNEx911T+gLDWzdnly26Wu1gevzUyS/yZ4V9G9Ojis/RnXIlctpVU/Q4+TJ8AhYGfr7+zE6KsgGEEPx0QeAY0cad1xzPQr3PazdJxENmQy6e4KAFyA8I0Xdfq2sGtOTwCWrgbNTQHEBKJWDQgK9/UBnN9rbOzAfkVUjVrbENeQGoqO8KxSPHgH27QlSTrV3AOs3AnMzyrpIiyS2F4e0zjVspUZ+vS3OzQYPi/m5oJDFxiuC/yqMZ6SdV8lp7xvAfHVWjaS2Ui7VZtWo+n8deyk98VhQ9KCevnXI3fuQU7anc2+R1XW1LcrqUfYc0XE147tqNbB1+3Ke6aSyw/SX5FptIDuHwrZJz8/6e8v4aO0zYk1XEHRXZ0srB9djLiSrRqRsQ/etCsv2cW46yIQ1uOFialkDcx+ImP/vuAmFex5UklWNzeeGaxSPHgF272wMAN6xKzxvfAwmdLdhwwap47hwFiCaGLmbb0PecFYNV2mlSWwa6i4Zrusv7R/WKriuO9eh/vRpFd3Zmv+toj/A/BorzYUzXTVEpFgAhRDiFz5/ZiSEJIPzPzlMR9espFQAhRDiGa74vBJC0ofzPzE+//jgwllEVAEUQkhLkx8YQm7HLuRuvg245nrkbr4NOVFhBkJIU8H5nxyW3G5CfP6MQAixT35gyHgVUUKIH3D+J8Pnktt84yzA588IhBBCCCGu4vPLSS6cBfj8GYEQQgghxFlYcrv5SLUACiGEEEIIcR4unAX4/BmBEEIIIcRZWHK7+aCPMyGEEEKIeXxeY3HhLIA+zoQQQggh5vF5jcWFswD6OBNCCCGEmMfnNRYXzgLo40wIIYQQYh6f11hcOAvw2f+GEEIIIcRVfF5jceEsYvMWoHegdlvvAGvRE0IIIYQkwGcfZyMlt3/+85/j29/+NkqlEm6//XbceeedNfsXFhbw+OOP4/jx4+js7MRnP/tZrFu3zkTTdimXo/8mhBBCCCFKtHTJ7VKphG9961u4//77sXv3bvzkJz/BG2+8UXPMj370I6xevRp79+7FBz7wAezfvz9ps/Y5sB+YGK3dNjEabCeEEEIIIVq0tI/zyy+/jKGhIQwODqKtrQ233HILDlSTo4AAACAASURBVB8+XHPMc889h/e9730AgN/5nd/BL37xC5Qdf3vr86ASQgghhLiKzz7OiV01xsfH0dfXt/x3X18fXnrpJeExhUIBq1atwtmzZ9HV1dUg7+DBgzh48CAA4JFHHkF/f79239ra2rTPnxpcj9ljRxq2rxxcj+4EffKJJPprdai7ZFB/+lB3yaD+9KHuktFK+pv9wEcw9dwhoFjlrlEooOsDH8FKDR2kqbvEC+ewN8e5XE75mAqbNm3Cpk2blv8eHR0NPU6G/v5+7fOL774NOPT/1uYZzBcw++7bsJCgTz6RRH+tDnWXDOpPH+ouGdSfPtRdMlpJf6V/+fvaRTMAFIuY/pe/x7n1G5XlmdDdhg0bpI5L7KrR19eHsbGx5b/HxsbQ09MjPKZYLOLChQtYs2ZN0qat4nNybkIIIYQQV/HZHTbxG+errroKp06dwpkzZ9Db24uf/vSn+MxnPlNzzLve9S78+Mc/xtVXX41///d/x9ve9jbhG2dXcGVQi0ePAPv2ABfOA6tWA1u3o6AYcaojozQyjKnvPo7i6VOBz9HmLcgPDEXLbu8A1m8EyiVg5SXBAdOTwb+uHqCrO9g2OxMpM21UriG3bkiq34vDJ1Hatzewl2o546PA4kKQoaW9A1jTBcycN6YfE/ZSGhkGDuxH+cyp2rGbmwVeOw7MzwEdK4GNVwT/VdCPrOyxK69BMZ/X0n3seM7O1Py/jK6X+10/ntOTwCWrgbNTtePa0y89ntKy83ng0jcp61yqna4eoKMDGD4RjIWi7diYQ4SQJqZyb5Dd7hC5soEovZ/97Gf4zne+g1KphN/7vd/Dhz70IXzve9/DVVddhRtvvBHz8/N4/PHH8corr2DNmjX47Gc/i8HBQSnZJ0+e1O5Xklf3pSceQ/mZpxu2526+Dfm779XukwrFo0eA3Tsb3EWwY5faA01RRmlkGOXdO4GR4YsbB4aQ27Gr5mEXKluFEJlpo3UNMf0ujQwjv+dPUTx9IlnnFPVjwl5Cx14VQb9tyq6gbZMRco30W9CGDdlh9z2tdiRtx8YcypJW+lxuGuouGa2kv+LjDwHPP9u44x03oXDPg8ryvHLVAIAbbrgBe/bswd69e/GhD30IAPDRj34UN954IwCgvb0dn/vc57B37158/etfl140Z4kTybn37Ql1F8G+PXZlHNjf+IBdelsVK1uFMJlpo3MNcf0+sD/5olmmnXpM2EvY2Ksi6rdN2RV0bTJKrol+i9qwKTtpO7K2Y2MOEUKam9kZte0OwcqBApzwcb5wXm27IRnSbioq/VBsKzU0ryGq3yavSUmWAXsx1fcwOTZlL5PAJtNwz6qXlZataLcjo08Lc4gQ0tz4nI6OC2cBTvg4r1odvr29w6oMaYMWyVYha3+mjpV650X02+TEV5IluhaFcTLV9zA5NmUvozuegHBMbY5naraiO89k7jW6Oh89HfhdE0JaDie+6mvChbMAJ34NffAT4dtLRfkHjoYMaYMWyVbh9Vcye3iWRob1XU2i+r15CwqDl+p3rMJAEEQlQ2lkGFiYD9+pMk6btwTtJkHUb5uykXA8AfGYmug3EN53m7KXKI0MB4GXmkTNz0Q6HzuD8u6dXDwT0oI48VVfk8JXvvKVr2TdiSjOnj2rfe6qVatw4cIFrXNLqzqBZ54OouQr5AvAR+9Gvj8lH+0f/hNw4jeN2+fnkDs3jdwNt9iR8f3/A3jjldpt5TJypWLt8SLZnd3AW98BDF0avO0sFID1lwFlNPovzVyQvxbDlP/ur4Hjxxp3dK0F3vxbwGVvDq7h3HTjojSi37nVa7D2d+/A7OjpIGtGlBwgeKv3prcAl18VHLe2F7m3vBW5rdulg6fKf/fXwCsvhu7LrWiX1m9u9Rrgv7wbuXPTQb+ixg4A1vYFGRSWjstd/XZhv23KXtaBzHiu7Q3yh0raYk2/q8dz1WrgwrnGXKQAsLoTuO6dseOpJTtGL/X3vfLf/TXw8guxclAsBhk1qpmbiZyf0joXXc+Fc5nNfxFJnhutDnWXjFbSX+lH/wKMnWncsaYL+VtuV5ZnQnednZ1SxyVOR9es5A49hbLo15Biei9dTPgs6siQ3S6UveHy0KjY4qMPAJNjDduz8nUUtrt+Iwr3Pbz8Z/HRB4CQKpJR/W4b2tCQfUUkB1dcXdOeDib9W/MDQ0BY30PGDoMblPpuU7bseEa1KZIR1u9lOWFjetmbpSPDlWWb0kudnOKjDwQp42TPj9on0rniPCKENCe5tb0IS+lGH2ePccHHOcqAZI1Lx+9R1k1F1Z1F2Jes/Jwlr1/Yb0UfTZvuPyZsRUe+U7IV7NmULXqhl1z4bV7azqN0koHOCSH+Qx/nJsQFH+fyrXcAK9obd3T1SPu+YvOW4Pgwonw6ewdqt/UOhPtnyhxXfXxPSC35DPychX6fousMcw9Q9dFU1ZcCQltp7zBzI7LY9zDZ+f5BJdlK41lp04QtpqwXVdnFo0eAo//ZuKO7N9zOFXSSmc4JId7js48zF84Csv41VBoZBp7cG+4T26boYVMohG8fHxHnUq2viyOqkyN7HJY+SV9+pVo/bHFgPzARkix94xUNfqj5pWIN6FvXeLxqPloFfckSaSvzc8CTe80sTCz0XSRLuS6TwngChm0xRb0oy963Bwj9IIpQO1fSSZY6J4R4jQtf9XXhwllA5r+GogoWqDxoRA+3JUKNNOycidHw4g0yx1UjSG6e9mQRtifoX35gCBAEhUr3XUdfsnKjFsYmik3Y6rtAdnnsjJJs1fGM2qdkiynrRVm2KMfy/Fz4dgWdZKZzQoj3uPBVXxcunAVk/Wsorh0TwYEAQv0LkwYHRrXpymTR6UfSvtuyKZnzbbVhYj6YkJ3FeALu60WYw1uwXUUnWemcEOI/WX/VTwIXzgKyvsHHtiMZUBMrJ8S/0FZwIAC7PqEKaE3aMF9nhVzLtmxK5nxbbbgSBJfFeEb10RW9YOt2IF93m8/ng+1hKOgkK50TQvwn86/6CeDCWUTGC7zQh1I1kgE1sXJC3D5kH4javxht+oRKojNpK77OuZtvA665Hrmbb0Nuxy75XMuWfmHHjrGtNgy9HQiVXVCTncV4CvtuUy+KsnN9A0F+6GrWdAXbQ1DRSVY6J4T4T9Zf9ZPAPM5RZLjAC80jXU1lwRuS/1VJDhoNVTaHtVau6yi/zZhrMYnupBXl3JXBVm7w2DG21YahvOahsotqsrMYTyADvajKPrC/MTfz9GTkfJPVSVY6J4T4D/M4NyM2g34kMOW3KvXrrc7tw6aPsyu/MrNwxaGPsz3ZWblWua4Xm/3L2p2NEOIv9HFuQrJe4JnyW5V6iNW5fdj0cXblYZvFpKWPsz3ZWd2EXdeLzf75/OAjhGQLfZybkKwXeOVb7wA6VooP6Fgp94AK89Wup97P2VYBFLjzsM1k0lrymzdmK1E4XgAls5uw4wVQbM43nx98hJBsyfrlZBK4cBaQ5QJvuaDF3OzFjYW22spwc7PyhS0kfLMbjNVCARTAnYdtZpPWsN+8cVuJwuECKJnehB0ugGJzvvn84COEZEvWLyeTwIWzgEwXeGEFLYqLjZXhZApbxBRAWabaz9liARRXHrbCySmZ5k8LG37zJm0lrh2HC6BkMp6A8wVQsvBxtq5zQoj3uPL1WQcunAVkucAzEfyjLKvKz7kVCqBg8xagp79xu2SaPx1s2JRJW9E515UguCzGE3BfL1bnW0Y6J4T4jytfn3XgwllApgs8hTc2cf2R7m+Vn3MrFEDJDwwBl1/ZuEOlnLkiNmxK6dwEbwJdD4LLYjwB9/Vic75lpXNCiP+48vVZBy6cBWT1GaE0Mgy8drxxx+qukApg8f2JLY5RfeySwbZCARQAwOxM6GZbE9eGTYWPb66x6AWQ6E2g6wVQAKQ+noD7BVACQRbnWwY6J4T4jzNfnzVgARQBNgsbRCLySV6xAjhfUu6PTAGUZUZPozQy3BIFUABxAnacPoniow9cfEM7dgY4cwrI5YE1ncDW7Sho2IANmwof3zLQtqLxYMmiOdLtOFQABVAYz9kZIJcDhk8EgZOrVjs1pkZlW55vac8hQkhzUL71DuDwoVp3DU98nLlwFpDVZwSh/OqsCTLHx+1fsQJYWKjdNnYG5d07gTXdUrJ8LoACIPhc/eIvGxcWk2PBvzDmZoDdO1HcsUv5wZ+qj7Omvai244ovLwC98QSAmfNOjalJ2dbnW8pziBDSHGT2ctIAdNUQkNVnBKH8VavVjo/bf91vA33rGrePDAPTE1KyfC6AAkT4aMZRKgL79iiflqqPs6a9qLbjjC8vEown4NSYmpRte76lPYcIIc2BUy/RFOHCWUBmqVJEwTxbtwMDQ7XbZQpbCOTlPno30D8Yfk7XWmsFUFwJDlxG4KMZy4Xz6udYuPbQ4icDQ4G91Gc86OnXb8vxAijL6I4n4MyYGpWdxnxLcw4RQpoCl16iqcKFs4BMU6WEBPPk+gaAu7bVLpJkC1sIgoOEBtqx0loBFO1zLKE9Sds79M4zeO2hxU86VgJ3bQvsJZerPaH+b1UcLoBSIdFN14ExtSLb8nzT1nlxkWnrCGlRmMe5CcnsM0JEME/u0FONvqtxhS2igoNEeViPH7NWAMVqwQgdNm9pfJMvifJD3/S1hxU/mZsN7OTA/iAYsJokacIcL4CyTILxBBwYU9Oy05hvujqfn0N5904unglpQXzO48zgQAGiaHHbnxFMBtzFySsMDKF4+ZWND9b5OSlZ3gcHIvDRLO3YBRzYH/ShOvNC5f9f/GUQQFbN1LhyZgLT127aVky3lYVsqfGcnQFOvQ5MT9ae7MCYmpadxnyT0vnYGeDka0CpLjNQ5Yd/yhl1CCHZ4tpaQAUunAVklipFUKSismBXXczH/gBQ8E8MCw403p8MyA8MRT64i48+ABw70rBddYKbvvY4eaEf5DWLoNgcN9Oy48YTWBrT+oUzsh9T07LTmm/SOjcwjwgh/uPiWkAWumoIyOIzgrD4yVIwj45PUNw50kZqqACKj35NpoIYTF97pDzD5ZC9KICigKtjalq2S/PN52AgQohZXLo3qcKFs4BMPiOIip8sLqD8nb3At/6H8mI+9gdAmH9ifYVCQTs6Py689GsK01HbCpRnZ5QWoaavXSQP+/YE9lJcbDxJ08/Z5riFyi5atomwMZXJUlNH6npRlO3UfDOkc0KI/zh1b1KEC2cBWbwdES7KpyeDT5yCggJJfIrzA0ON2Trq/RAFsprBx1mG/MAQcjt2Ae+4KSgcAwCLC8DzzyoFNxn3cT5zKnzH2JnAXkJcEXTbc92XV5VQu5fNUlOF63pxab6Z0jkhxH9cujepwoWzgCw+I5guThG1r3p7aLYOCVm+F0BRIT8whNzKSxqrLcZlNanC+LULFsa6/dA5x5VCH1rt6mSpqZfhuF5cm28mdE4I8R/X7k0qcOEsIJPPCJu3AG2K8Zq5XPRiXqIAgtQvvGYtgKJA4l/Ipq9dUB0wkkJBrz1fCqAoYOSNBwugKOHzWyZCiDno49yEZHaDrzekOMrl+E+dMQUQpH7hNWkBFBWM/EI2dO2lkWFA5KoRRSFBIh0PCqCoYOyNR4sXQFHB57dMhBBz0Me5CcnkBn9gvzCHciRRnzplCiCIMjBEnSMrW6c/jpL4F7LJaz+wX8q9poH5Of32fCiAooCRNx4sgKKEz2+ZCCHm8PnrExfOArK4wQsNZnUncM31QXDa6k6lc2W25weGgMuvVO5fqwQHVkj6C9nktQvPuezNyN18W2Avirai054rQXA6mHjj4bpeXJtvPr9lIoSYw+evT1w4C8jkBi8qfvL2G1C472EU7nkQubffEH6MohE2bJcohNLKwYFA8kWI0WsX2cqlb0L+7ntRuO9hoa3oFEFxPQhOBxOLStf14tp8c20hTwjJBp+/PnHhLCDtG3xc8ZPl9hWNTfb42AepoQIo2LwF6K5rq7s3k2Cl0sgwSk88huKjD6D0xGOxKbGSLkJM3ShkbcVkERQfCqCkPZ6A+wVQdIIDVfSYhc4JIf7j89cnLpwFpH6DFxU/2XhF4EpRaV/R2KSPj/NzNlQApTw2AkxN1G6cmgi2p0hpZBjl3TtRfuZp4NgRlJ95OjYnc9KFjLEbhaStCF1wNIqguF4AJYvxFPbdoQIoAJSCA1X0mJXOCSH+4/PXJy6cBaR9gxcaS50LRRJf5qjtMn7OJnycsW8P0FChvry0PUUO7A+CKquJySfrio+zrK0Itxls0xVf3izGM6qPTulFJThQRY8Z6ZwQ4j8+f33iwllA2jd4WSNK7MsctT3Gz9mEjzMunFfbboksAhtN3ShU5GTRpipGXCYyClRtNr2obG+14GBCiDl8/vrEhbOA1G/wsr6Im7cAVZ/jAQAdK8XGpuDjGPlANlUApbrcbjU6xTwSILxWQeBcaWQYGD2tJqseU8UoVOSEueD09NttUxUDBVBUxzNqn9Ki1/ECKMp6UdBJZjonhHiPz1+fuHAWkMlnBAlfxPzAEHDXttoF6NxsdBEUWR/HKD/nxcVksrG0+KyfKACQzwNbtwvPs4JC4FzFlxNjZxqPHxhSWySZKkahIieXi/7bRpsJZSsXQFEMhJQOsJTB5QIoinaupJMsdU4I8Rqfvz5x4Swg9c8ICr6IuUNPNRa/iPJDlJQb6ec8PZG8AMqB/cD0ZOP2q9+OwrXXh59jCaXAuTBfTgDoW4fcjl01AXmRmCpGoSLnwP7gmqrRCA50vQCKciCkZIBlLI4XQFG2cwWdZKZzQoj30Me5CUn7M4It30LlX3URfs5JgwOFbWZVAlgycE7Y7/5BpQe+7eBAU36oSdtUxZhshUBIHfuXlR0pPwvZSe08SicZ6JwQ0gTYdHOzTFvWHXCV1D8jKPoWhi41Q2SIjo36tSdaxoYFB4Yee/I1FB9/KPj/6cngX1dP+NumiL7YRtj/U6+j+OgDF/X5xqvC8020p3z9lmwlClN9Ly19GSmfORVrFzh9EsWv3bd8XG5d4BYj+rEiPZ6zM8Dpk0IZKpjQy7JOJsel+mhUL1VydOantM6nJ4ETv1GWH0bx6BHgiceAc1NALg+85Trk7vrvfGtNiG/YdHOzCBfOIhSDZ5Kg5Vv44i8bH3RLvoXVD5DyrXcAhw/Vvj2PcjnZvAV44flGl4p8PrwASr1sADg7BTz/bO22MP9gINtfmKJrrSz2o9Dot/JYhGDTVtLoe3n3zlq3F5FdAMDkWPBv6bjyK8eA48dQErnHJBlPIJMxDdVJHDb0ojs/k+pcMUagePQI8NiDuJjSsggcfR7lP/siSl94hItnQnwhyhXt7nuz6ZMkdNVwAYu+haouJ/mBIeCKqxt3lEpyBVBUydC/MT8w1OjHLotGv424/9j2QxVgrO+KFQsbiMgTnGg8gWzG1IROAHt6idFJItmqMQKAIA88gKlxM37lhJBUYHBgMyLyu7Pgj2fTt1DLOJP6RKqQtX9jfZClLBr9NnGjsO2Hqtqukb4rEilHdzyBTMbU5EPCil5kdKIrWzFGAEBkvncfHriEkAAGBzYhaQ6qTlu2CqaYkK1C5pNEM3+0Tr9N2JTN8TTdbpJjteUkyAeexZiatH8bepHqX4pzKKqtzO8lhBBpWAClCUl1UMOKmsT5/smek7ZsFVRzINtg63b1T826/dYZCxMysmpXRoYqcW3qjKeMXBFJ9WJCJzJt2rTzNOfQ1u0AQvKQd/dmfy8hhEjjcwGUXFm50kC6nDwpiCyXoL+/H6Ojgoj9GEpPPIbyM083bM/dfBvyFhzXqyPrc2t7I6PkVc/Rld3xw3/E7OlT0rJrMgJUR9NPTwLdPUBn9/J+2X6kQfHokcB38sJ5oL0DWL8RKJdCryE3EJ3BoILI9nTGwoSMrNoVyqhk1ajYxdxsEPQ4Pwd0rETblddgMZ/X0n3seFbbpwFbTKqXyDlUpxdsvCL4b4ReRLYXqpe5GSPz08YcimzLYlaNJM+NVoe6S0Yr6a/46APAsSONO665HoX7HlaWZ0J3GzZskDqOC2cBpgfVR1ppEpuGuksG9acPdZcM6k8f6i4ZraQ/0y8n01w401VDgM+O64QQQgghrkIf5ybE50ElhBBCCHEVn32cExVAOXfuHHbv3o2RkREMDAxgx44dWLNmTcNxH/3oR3H55ZcDCF6nf+ELX0jSbCrkDj2FsmhQr70+m04RQgghhHiOz3mcEy2cv//97+P666/HnXfeie9///v4/ve/jz/4gz9oOK69vR3f+MY3kjSVOj4PKiGEEEKIq+TW9oaVMvLCHTaRq8bhw4dx2223AQBuu+02HD582EinXIA+zoQQQggh5vHZHTbRG+epqSn09PQAAHp6ejA9PR163MLCAr74xS+iUChg8+bNuOmmm4QyDx48iIMHDwIAHnnkEfT392v3r62tTfv82Q98BFPPHQKKVe4ahQK6PvARrEzQJ59Ior9Wh7pLBvWnD3WXDOpPH+ouGa2kv6nDT2M2xB125eGn0X3r7ynLS1N3sQvnr371q5icnGzY/rGPfUy6kW9+85vo7e3F6dOnsWvXLlx++eUYGgrPublp0yZs2rRp+e8k6UUS5XH+l7+vXTQDQLGI6X/5e5xbv1G7Tz7RSqlxTEPdJYP604e6Swb1pw91l4xW0l/x9KnQ7bOnT2FBQwdppqOLXTh/+ctfFu7r7u7GxMQEenp6MDExga6urtDjensD94bBwUFcd911ePXVV4ULZ1egjzMhhBBCiHla1sf5xhtvxNNPBwmsn376abz73e9uOObcuXNYWFgAAExPT+PYsWO47LLLkjSbDpWKV7LbCSGEEEJIPJu3AL0Dtdt6B4LtjpPIx/nOO+/E7t278aMf/Qj9/f343Oc+BwD49a9/jX/913/Fpz71KZw4cQJ/8zd/g3w+j1KphDvvvNOPhTMhhBBCCLFDfeFqtwtZL5No4dzZ2YmdO3c2bL/qqqtw1VVXAQCuueYaPPbYY0mayYbZGbXthBBCCCEkngP7gYk6n+SJ0WC7RsntNGHlQAFMR0cIIYQQYh6f48i4cBbgc45BQgghhBBX8fnlJBfOAnyuo04IIYQQ4io+v5zkwlmAz58RCCGEEEJcxeeXk1w4C/D5MwIhhBBCiKv4/HKSC2cBPn9GIIQQQghxFZ9fTnLhLMDnzwiEEEIIIc7icQEULpwF+PwZgRBCCCHEaTwtgMKFswiW3CaEEEIIMU9UARTH4cKZEEIIIYSkhs9f9blwFsGS24QQQgghxmFwYBPi86ASQgghhLiKz5nLuHAW4POgEkIIIYS4is+Zy7hwFuDzoBJCCCGEuAp9nJsQnweVEEIIIcRVfHaH5cJZBNPREUIIIYSYhwVQCCGEEEIIkYQFUJoMpqMjhBBCCDEPC6A0Hz773xBCCCGEuIrPcWRcOAtgOjpCCCGEEPP4/HKSC2cBTEdHCCGEEGIen19OcuEswOfPCIQQQgghruLzy0kunEUwHR0hhBBCiHF8fjnJhTMhhBBCCEkN+jg3I0xHRwghhBBiHhZAaT58/jVECCGEEOI0LIDSXPgc8UkIIYQQ4iwsgNJ8+BzxSQghhBDiKgwObEJ8HlRCCCGEEFfx2R2WC2cRTEdHCCGEEGIcn91huXAmhBBCCCGp4bM7LBfOIpiOjhBCCCHEOD67w3LhLIKuGoQQQgghxqGPMyGEEEIIITKwAEoTQlcNQgghhBA7sABKc+HzZwRCCCGEEGdhAZTmw+dUKYQQQgghrsLgwCbE51QphBBCCCGu4vNXfS6cBfj8a4gQQgghxFV8/qrPhbMIpqMjhBBCCDGOz1/1uXAmhBBCCCGp4fNXfS6cRTAdHSGEEEKIcejj3IzQVYMQQgghxDwsgEIIIYQQQogkLIDSZNBVgxBCCCHEPCyA0oTQVYMQQgghxDgMDiSEEEIIIUQCBgc2I3TVIIQQQggxD4MDmxC6ahBCCCGE2IHBgYQQQgghhMTA4MAmhK4ahBBCCCHGYXBgE+Kz4zohhBBCiKv4vMbiwllA+dY7gHyhdmO+EGwnhBBCCCF6MDiw+cgdegooFWs3lorBdkIIIYQQog+DA5sLn/1vCCGEEEKchcGBTQjT0RFCCCGEGMfnl5NtSU7+t3/7N/zDP/wDTpw4ga997Wu46qqrQo/7+c9/jm9/+9solUq4/fbbceeddyZplhBCCCGEeEpubS/CHDOaPjhw48aNuO+++/DWt75VeEypVMK3vvUt3H///di9ezd+8pOf4I033kjSbDowHR0hhBBCiHF8TsCQaOF82WWXYcOGDZHHvPzyyxgaGsLg4CDa2tpwyy234PDhw0maTQe6ahBCCCGEGMfnBAyJXDVkGB8fR19f3/LffX19eOmll4THHzx4EAcPHgQAPPLII+jv79duu62tTfv8ifZ2zIdsb29vR0+CPvlEEv21OtRdMqg/fai7ZFB/+lB3yWgl/Y2fP4uFkO1t58+iV0MHaeouduH81a9+FZOTkw3bP/axj+Hd7353bAPlkPQiuVxOePymTZuwadOm5b9HR0eFx8bR39+vfX5xeip0+/z0VKI++UQS/bU61F0yqD99qLtkUH/6UHfJaCX9lVZ3hm5fXN2ppQMTuovzoKgQu3D+8pe/nKgjfX19GBsbW/57bGwMPT09iWSmAl01CCGEEELMs3kL8NKvgPGRi9tYACXgqquuwqlTp3DmzBksLi7ipz/9KW688UbbzRJCCCGEEFdpxQIozz77LD71qU/hxRdfxCOPPIKHH34YQODX/PWvfx0AUCgU8MlPfhIPP/wwduzYgfe85z3YuHFj8p7bhlk1CCGEEELM43EBlETBgTfddBNuuummhu29vb340pe+tPz3DTfcgBtuuCFJU+lDVw1CCCGEEOP4XACFlQMJIYQQQkhqiAqdNH0BlKZGkFUDZwXbCSGEEEJIPJu3BMGA2cEnRQAAIABJREFU1TA40HOmJ8K3Twm2E0IIIYQQOVoxOLCp6Vqrtp0QQgghhMTjcXAgF84iuHAmhBBCCDEOgwMJIYQQQgiRgMGBzQjzOBNCCCGEmIfBgU0I8zgTQgghhNiBwYGEEEIIIYTEwODAJoSuGoQQQgghxmFwYDNCVw1CCCGEEOMwOJAQQgghhBAZGBzYhNBVgxBCCCHEDgwObDLoqkEIIYQQYh4GBxJCCCGEEBIPgwObkemp8O1nBdsJIYQQQkgsDA5sRqYnwrdPCbYTQgghhJB4GBzYhHStVdtOCCGEEELkYHBgk8GFMyGEEEKIeRgcSAghhBBCSDwMDmxGmMeZEEIIIcQ4PgcHtmXdAWdhHmdCSAoUjx4B9u0BLpwH2juA9RuBcunivWZ6MvjX1QN0dQfbZmeCB8zmLcgPDEm1UxoZBg7sD97ohMnu6ACGTwBzs8Cq1cDW7Shce7183yXPqe6H6jXotmm6D9J9PDcd+G0ObrgYCGVoPG0QaotzM6H9Xbahqr6jvz+5bMP6MWEvFWzMIZv9zmJ+SrN5C/DSr4DxkYvbPAkOzJXLbntjnzx5Uvvc/v5+jI6Oxh8YQvHxh4Dnn23c8Y6bULjnQe0++UQS/bU61F0yWkV/xaNHgN07gVJRT8DAEHI7dtU82MJ0VxoZRnn3TmBkWF52vgDs2CV80Ib2Peac0H6EXIMInTZV+5DU9hKNqYIuTGPCFvt2PY7JtnbzsqvaUNGPCXupYGMO1VOxPRP9zmJ+qlAaGUb5G/fX+jn39CP3+a/9/+2df3BV1dnvv+ecQEIgCflFYhSrMlpvq2+tglLGkXbKMMz9o/Qyd265Q/XS1r5FgQqCP6oVUeTFKYhlgNdWtAK+1Oud22GY27mj1HbEYWpf6fi20FcFA1h5gUB+kBCSHJKcs98/Tk7IOWevvddae6+990m+nxmGZO+1nv2cZ6919pO1n2c9Wtfy45nR1NQk1Y6hGiK4jzMhxDQ7t3hzJoZWh1zZt0ftgQ9k9Nq5RXzeTne3PnZ6yH4G3Wv6rYMbXu6pn3qo4sNY7HnjZTOyR1xDyT5+jJcsJuaQCD/0DmN+qsDkwFEI93EmhJimt8ezCJlkGu2EGyf9ROcc+nhOCNK4pu86uOHxnoaWHOXDWEx1CFb8fJCdRck+PowXret6vJYveocxPxVgcuBohNvREUJMUz7RuwyZvAvd3IzxpeJzIt0d+nhOCNK4pu86uOH1noaVR+PDWIxNKDcme/gaKveptMz+uI4+JuaQCB/GeSjzU4FiTg6k4yyCjjMhxDT/7V7vMk6dzCT0CEi3tgCfn9AWL5Qt0j2dEvax7p6bibMcSTyROS6DxjV918ENr/fU5X4aw4exOHjyU3vd/RjnAFDfKJ08lm5tAQb67U8q6mNsDonwYZyHMj9VYOVAQgghqsSO/Nn+REUV8JU7M/+uvwmonQLc8EVgcm1h245W57hAu1hCICNrpGy7RYGuDqFsoe7dXeI+B/fbxl3GDu4X6+/xmn7r4CpfpGNVjT/30xBCvcsnXdE3Oya/eKut7um2c7a6K8keaZ8R14vdNVstSW3fHqCnW+2zOskyMIdE+DLOQ5ifyhRp5UBuRyeCyYGEEMMI4/marrXdvSe16Umgs11ejtO5hiYkVq/PlX2xU7q/zjW9xjVqfU6fddCVj8arC+6pzv00hfCaU6/PGSdZVHRXle0HfowV1/Ye55COfibnSqBxx07Jgfev8v96PsIVZxFMDiSEGEY1zk8Y/+cUfxmz/5rPl6Uq2zEWUbGPbFyjUzuvMvyKrVSRr3U/DWFyLIYRz+rHWLnSwcwcEl7O4Dh30ifI+8TkwNEIY5wJIaZRjfObvwiotikyIYiLTX1yBPjkcGH7qprCayjKxvxFmaIPdjj18RDXaN09FxhXuE8wxpfKx2Gajq1Uka9qc5OYHIthxLOKxqfKWIHhOSRAOM4rq+VtFsL8VIHJgaMROs6EkCBQiPOL1zcC195QeEIUF7tzCwB7efmxosqyASCRsD/u1EczrjHd2gLs3mqf8NV/Gdi9Vd45MR1bKSlfy+YmMTkWw4hntRufqmPF9BzKw3GclyhG1wY4P5VhciAhhBBldIoAJPtsD9u+4hTt2dp/2btsUcKUah/ZogduBShUisGYLLygKl/F5iYxORbDKHbhND5VinqYnEN2OI1zlT+ogp6fOhRpciAdZxFMDiSEGEYnzk/pFadoL1fBcRXZro6ATRyll7hG021MJweKjkfllbXJsRhGPKubbOlrG5xDdvild9DzUxlWDhyFMDmQEGIYnYes0l6rix8C4nlf8/F45rgdCq9PXR0BmzhKL06F6TZhJAcCiMwra5NjMZQ/DtwS8mQT9hTnkNe9kF1tIql30PNTFSYHjkYY40wIMYzOQ1Zlr9VYbT0wqTL34KTKzHGhUnKvT211H4nNa2UvToXr9bzI8bHIg5b8CLyyNjkWAy2sAcmCJZIJe6pzyOteyK7jXFLvoOenKlF506IDHWcRdJwJIYbRecgqrdTs21O4r+zFTvHrUIXXp7a6u+jkxalwvZ4XOX4WQFGVH5FX1ibHYuCFNVziewHIxwsrziGvK6mu41xS76DnpzIRedOiAx1nQggJCdMxzn4VOlAqajGSvNfKjHEOXh9ZRlOMs29xwA5tTMWsBzkX/JyfWkTgTYsOdJxFMDmQEGIY0zHOfhW1UCreMZK818qMcQ5eH1lGU4yzL0VCXNqIjhuPcfaxjZ/zU5mIvGnRgY6zCCYHEkJMo/G6UuV1qupDXKm9ne755L9W9vB61rp7LlBaJm5QWibnnESpAAqCj/8VYnIsBv1aXmZsVtfJjzuF++NLjLOpcZ6Pj/NTlai8adGBjrOICYItaETHCSFEB8XXlSoPHNWHuPJDX+LVaoFeGq9nh4tCXE5eOZgoya2udjkpX9giIgVQgBDif50wOBYDfy2fLz8Wc/5dgOr98eIQDracMTvOZfQK6D5F5U2LDnScRfQJNj0XHSeEEFV0XlcKtqMKOsZZKgELyNVX9/WsXVGI1GBhdTWZwhYRK4ASmZU3k2Mx6NfydtfLdwAlk+yCjHHueeNls+PcDj/mpw5MDhyFcMWZEGIY1YeycJstxb2W/Tgu7diNiKPUdRJVnEhdWSyAYm4sFnNyoNa+3HmluFHfKOUQpjokHN0hfJszPsxPbZgcOMrgijMhxDDKD2XRKtLU6xHPf1gD6qs6fhZAyTJiZU/XSVRxInVlsQCKubEY+B8HssVNZNop3p94fSNiK59F7K7ZwBdvReyu2YitfNZ+fuYRU1iY823O+DA/tWBy4CiE+zgTQgyjnLwnWvlJ9jlcRHFVx68CKCPbDumtmwgnvFYsv6KbpiwWQDE6FoNMgJQqfpJFspiI6v2J1zcifv8qJFavR/z+VVJOc7q1BQMnjhaemFhpU7nQw5yxa+txfuoQmRAlDUrCViCy0HEmhBgmdnA/LJvEI7z8M6Ru+GImEejzE0D/5UymfcL+Qai0Kphd1bl/laf2trqLGFrZE33e2MH9wM23CrsLr2Wl/ZEl0U8WZfmq98gQwrH40gakxo0HBgcyDuP40syOFIL4WdutC93GOZBxuLMrwMm+jJz5i6SczhxEK+GVkwsLmWRXW53sHNT92bcHVvv5wuPjxgE9Po1zER7npw6xyTWw+/OjGJID6TgTQkhIWOcFq13dXcBfP8g9Njhg39Zpq7MwCqAkEkAq7+E7tLLHGOfg9ZFFOBZ7LxX+3tlu2zRe1wDLZiwqjfNsHwA4cRRpyTCH4X4iuw3Yz5+wx4urvJE7bChcP8j5qcX8RcCnH2X+eMnC5MAihwVQCCGmEe0Xr4IovhkhFUCpsHkrN7Syxxjn4PWRxoexWHL9jfZjUVe2zO4ReQjtVm4fPxz2eHGV57PeJuanNhEIUdKBjrMIFkAhhJjGj9Avp/hmg8mBwrbVdbairc4O7UQ426IQ1XWFsmSKQ0SsAEpUkgP9GItWX6/vslVXO23HSn0jsPihwrEpUwQlqPszfxHidQ2F11n8UOEuHV7GuYH5qQWTA0ch3I6OEGIaH5wV19UgQ8mBwraVVbZNh/VU1Me2+ElpGfD9lcD3VuQ6SbLFISJUACUQfWTwYSwmauydMi+yVVY7hWPlvuWI1dZrF0EJ6v5YNteJ1dYD9y33b5z7PD91iUqIkg50nEVwOzpCSNRx2x9WdVVHpb2obVavkWRXyHRWmeyKn1xOInZwfyZpKT8G1O31fsQKoBTzylsO9Y2Y+D//0XeZSqudDmMF+/bkxtMCckVQHO5PurUF6VdeQGrTk0htey7zL+/n9CsvyO3cYZccOHQdX8c5UDg/S8bBSvbBevOVwMZiZEKUNGByoAiuOBNCTCMKs5gwEbjpy5mcivwtqsaXAtdch9iQU+GUOBVKcmCyL7NCtm3dlYf90AqZNcl+tctplUlnZcpveSoUa3KgcCyWlmVKPvd0564+jhsPTPsvgJUe3gGjpLEJaLPZ0cJtnGfbeNxVw++x4ijz/FngxTWFjnp+O0AqyTGocY5kH2Irn804yR/9WyZpcnAgk6RZMk75OtoUcXIgHWcRXHEmhBhGuCXTP0xH/P5VSL/yAqx8x7n/MmJD+8Rqy3dY7ZFt79j24H5Yditk6bRND5dVJpeyzqpbWpneBsukzU0i1OO2uwAA1r8eyD0x0I9Y1WRv43BonPuFmy1tgw5ciqCIZOJiJ2C3fZwd2dVhp88a4DiP1zciXTYBVv5OI4Kde5gcmAtDNURwxZkQYhi3ggOeVyMNJqo56S7Ub0K5UoEFt7LOOgUbIlcAJSLJgTr3U7q0dUCFNRyvM3+RfWKcWxEU0f2prFbTzcFWYYxz6e8QQwVQijlEiY6zCK44E0IMEzu4P1MIYiTZggNwWOmRLSkMGEtUc9JdqPelbsfPW4ComMXgAKxdW4GdW9TkuejtB1ryI7DypnU/JcehaZu7XQevbs6Ml9RgYSeZOGeFJDuhbk6rtm7j/NXNvo9z6VVkA/cJiFCIkgaeHOf3338fDz/8ML7zne/g+PHjwnZLly7FqlWr8Mgjj+Dxxx/3csng4IozIcQwrg8P3VWyLAYT1Rx1F+l9yX4ffOWH6MVO4OgR4atyy8E2UYtxjsrKm9b9lByHoRcR6WzPjJf8yoEyeqgk2YlwSXJ0HeeCgjOeYvlF91TxOrqM2eTAqVOnYvXq1Xj55Zdd2z799NOorKz0crlg4YozIcQwbjGZ8fpGpK69ofDBLVMqGGYT1dxiKG31HrRZ8YN6/K8rDvvtRy3GOSorb1r3U3IcBhXHrTtenPRwS7LDvj2ZNiMSGxGLAS2nM0mx5ROB+5Y7Jgaa0Fv7u0XxOtoUcXKgpxXna665Bk1NTX7pEi244kwIMY1MfKtgRwIZx8poFTs33UU7KZTkrde4xVznt5dhQrn4XMQKoERm5U3zfko5+AHFcVt3z5XfmzlLIuGoh9P9iQ8l6SZWr0di2U+RWPZTxP7X8sybkM72zEJb+3n3PZd1xnks5hx77OG7xbGPn0QgREmHwGKc169fj8ceewzvvPNOUJf0BlecCSFB4PLw8BJfqppUpJyE5KC72PHLc2zcHpb5+sjQ2uLsqESoAEpQiXNS6NzPtnNyYUOGbT5c/ERVbsLZYVW+P3Z7ScuUDlcd55bl7pDrfrc4yfCLiIQo6eD6J866devQ2VkYF7Rw4ULMmDFD6iLr1q1DTU0Nurq68Nxzz6GpqQlf+tKXbNu+8847w871888/j7o6uRgcO0pKSrT7t1ZUIm0TPxevqPKkUzHhxX5jHdrOG2PFfl2vb0PS5uFR+tb/RdXKtQCAwcXL0dH8cUFxhPjpv2PyYH9m79wRjLRd16EDSNokCJUdOoCqu79RqI9CezfdRXoXbHmV93kLrtF/ueB4DrFY4cP9ctJZpoPeXseezD3Naa94j0whcz87P2tG6tzp3Dbt5xHf8gwmr90itJ2qTbT1l3Hg8+m/7KiH6v3p6OmG3aZuJT3dqBGMK6lxbkdri/Y4B8TfLU59/ELHTk4E+cxwdZyfeuopzxepqcn8VVNVVYUZM2agublZ6DjPmTMHc+bMGf69zW4zdUnq6uq0+6e7LwqOd3nSqZjwYr+xDm3njbFiv9S5s7bHk+fOYiD7+UvGw7rmuoJEuHTbOXTs3FqwD+5I20nJV9VHtq1Abz/0wcSKTBGYyTWwznwOnDrpWWa2vdexZ9LmJpG5n+mHngZe+GnBPU2dO42OnVsx5fENtrYL4jMKx0r5JODGL2XCEv7js0whFwU9VHVPT6ywbT84sUI4rtzGOcomAM0fK+nu5btF9hpe0LGTE348M2RDj42HaiSTSfT19Q3/fPjwYVx77bWmL+sdxjgTQgwjHd+qGV9qMsZZqq1MDKWDLGFRiFtuR2L1esTvX4VYk+B54lJQQloHRYzGlRtERo94fSNQ12DbzmksBvEZhde49Y5M/PHq9Yjdcrt9Z4ewJ2Xd5y8q3G3DrXS4yzhPLPupUHfP40pijhpLDlS1U0Tw5Dh/8MEHWLJkCY4dO4bnn38e69evBwB0dHRgw4YNAICuri6sWbMGjzzyCJ544gncfvvtuO2227xrbppL9ivOwuOEEKKIbPykdpyzTgGUqrxrVdUoF0Bx1duhTxa3ohA5OitslRbJAiiSNjeJybEYRBy31DV0ttVTnEPx+sZMyfnaKZmFttopjrtqyI5zU/kKrnPUULy9qp2ihCfH+c4778QvfvEL/PrXv8aOHTvw5JNPAsiEZvzkJz8BADQ0NGDjxo3YuHEjNm/ejAULFnjXOghE8UY6cUiEEGKDdGEIL/voqiSqtbcWbuXWdSFzPA8p3WX2ihUVWBAVhZh6fc7DNV7fCFx7Q2E7QWGLqBVAUbG5SfwYi4MtZ7zJ9oDMNVTHyjAKc2g4SbH9vNyuGpLjXNWGnu+nxDW8oGynCMHKgSJEW9rEFbe6IYQQAbJ7+Go/8FUz13duAQp2f7WGjjvraHdcqLeELKf9c6WOKcoNrQCKgs1N4sdY7HnDvqZDEHtVS19DNexJp4iQwq4asuPc1J7sMnPUyJ7iuruPRAA6ziJKxtkfTwiOE0KIIkrxiRpxzsoOS69gu02b46ZjKH2Pt9Zoq4OyfAWbm8SPsZjqsE/OCjXGOe+4qi4miwip6GM0dt5ljpqIcY5K4R8d6DiLqBDUoRcdJ4QQVRTiJ3ViS5X7lJbZHy+3SYqW1N31oSuKF1WJLbV73Vxd512uDqryVWxuEh/GYkxUeCaAAijW3XMLbWmXbGaXlFZaJozjVZ5DikmppvT25bvFoY9XopIUqwMdZxFMDiSEBIFs/KRuYpNkn3RrS2FMJADE48Dih+zly+hu98AfiaAMt7T8LPnhdU4V5CJSAEXL5ibxOBYHT34qHosGbT4cL3s5eeVgaZltstlwUtpIZ/VyUhxfqzqHZBJag9Ab8P7dAjjPTy8EVE3SBHScRTA5kBBiGoX4SZ04Z6U++/YAFwuLXeGmW5C4+VZt3eP1jYitfDaTNW/HxQv2+qvElu7bk/lMIxHZxXTFMlW9VWxuEh/GYrrtXHg2z3ceLyedE+dGOquAML5WeQ5JJPoFobcv3y2AeH76AUtujzKYHEgIMYxynJ/Ofs6SfYQyBA8zFd2d9v8V9VGRb6qtDn7oEoYDYXIsRsnmWvp4nUOK89aP4359Rsc+XhjNJbfHLEwOJIQYJja5pmA/hexxlfZucc62fc6eQuqfVmdWPCur7VfKNHRR1l3URyFWVMUuwranTiL1+P1or61HKlvo6mLnFftUDuW3JPsyOsxfZLuKKJTfchqpbc/lylS0uUlMjkU3m+fYV8PmOsVufPmsUZlDCjb3bX56pJiTA+k4i6ioArpsbiCTAwkhPmHdPRc4dDA3ztWtWMaxfy98QA/FW9o6FfMXAR//tTAkIOugAOKSuw4xh1q62+kRjxf0UY0VVbGLrd4A0HsJ6L2EQTtb5B2zAODEUaRXPltgc6H8rg7grx8IZbp+RsOYHItuNre1haTNlceKSB+dsRvGHPJic5/mpx+oOvZRgqEaIpgcSAgxjGpRA+045/wKYrKIYjOhqfv1NxWeSKcL+yjGiqrYxVZvHQSxpZ7lO9jcJCbHolGbq8YVi/Rx+6xRmUNebO7X/PQDJgeOQkTxPhJ7khJCiAxaryt14pzzE4pk0Yh7NBpv7fT961W2Bkpxy7KE9IwJ5X5q4MdY0fqso2QO+f7d4gUmB44yBgfUjhNCiCI6e5lq7X+quSew33qYKvbgh2wdVAqseJEZBMVqc84h++OB2UWXIk4OpOMsQpQcOI7JgYQQn7Db49iu+IHXPosfUn/VbEIP2T5By9ZBpI8X+W6f0STFanPOoXDtokkxJwfGLCvaa+NnzpzR7ltXV4e2NvssVzdSD98LdHcVnqioQmLz69o6FRNe7DfWoe28MZbslx6K27Q6O5x3DpDsI7Jd6pMjwM4tmVLO40uBq6YCl/syyUBV1VcSn912MDCoexiyh3ckGEr0KqmdgsFsBbxs8peifYblnz9b2F9TZhB4tXlZw1W4PO+/K9vcq31MjpWRmJ5DTvbT0T0ou+iQfuUFWP96oOB47K7ZiN+/SlmeH8+MpqYmqXZ0nAWkli8Ekr2FJ8rKkdj6v7V1KibGkvPiN7SdN2g/fWg7b9B++tB23hhL9ku3tsDa9GRu4aKaesRWr9dy1IN0nBmqIYIxzoQQQgghZmBy4CiDMc6EEEIIIf7D5MBRiMhBFjnUhBBCCCHElWJODqTjLGJAEJIhOk4IIYQQQlwJdOs7n6HjLGKgX+04IYQQQghxJ8Ct7/yGjrOIdFrtOCGEEEIIcSVe3wjctxyonQJMmJj5/77loW7HKAsdZ0IIIYQQEhjp1hZg91ag/TzQ15P5f/fWzPGIQ8eZEEIIIYQEx749QL6TPFR8JerQcRYi2k+wOPYZJIQQQgiJItxVYzQi2oi7SDboJoQQQgiJIsW8q0ZJ2AoUI6l//DYQjwPjxmf2e06nMw51InHl5/GlwKRKoLsrU21w5LG+HqCyGqgcqmt/sTPzb8LEzLns/3ZtRh5L9gFlE+TOa/x8obIKqf5+INmH2OQaWHfPRezg/sxfhII++e2c+vnZVuf34ezdfXt87YP5izA42I/0zq2e5eS3KdZj8frGTOyaZNvBljM59vNLrl1bkzLCOD64eDlQMl5ZDgDfdNKR5dQnSHl2Y8+LvKidMynXhO1M9g1Ttt357Nw1JT9S5+cvAj79qKDkdjHsqhGzrGgvoZ45c0a7r5fa5akffkv7uqOWeAJIp9TbOfXzs63q79V1QCyWO3H96FNdh3gigXTbOc9yCtoU47Fs9vTurbkxbQ5tE//yz0idO+273IK2JmWEdDzRcDVS331QSU5s5bMAAOvFNZ510pHl1CdoeQVjz4O8qJ3L/qFhSm58yzO+2s6Lvm59w5QtOp9ouBrph542Jj9y5zc+kVs9sLoOsUf+SWtnDS/+Xpampiapdom1a9eu9XQlw3R3d2v3LS8vR29vr1Zf6/+9oX3dUYvs31gq9ef9bKv6e7IX6Ot1bqPTJ9kLq7fHFzkFbYrxWO8l4NjfgJF/SLi0tfKTRnySW9DWpIyQjls93cpyYpcuAkePAMf+3bNOOrKc+gQtr2DseZAXtXOx22fBeuOXxuSmjx6JjL5ufcOULTpv9XQblR+5880f555P9g6fV8WLv5eloqJCqh1DNQgh5sn/QyJKbU3KKJLjjgk5Achy6lPM8qJ4Ticpq1jluvUNUzbPF29yIB1nQoh5yofi9qPY1qSMIjmeTcixfd8TgCynPsUsL4rnYpNrxoxct75hyuZ55/NRhrtqiPiv/yNsDaJHPKHXzqmfn21Vf6+uyyQj+N2nug7xugZf5BS0KcZj9Y3A4ocKy6s6tE00XG1Erm2JV1MyQjqeaLhaXc78ReISuEHIcuoTsLyCsedFXtTOAUbl+m47L/q69Q1TtuB8ouFqo/KL6nyEYXKgA6m9/wL8//+TezAWAxDL3VXDSgPpoV01sj+P3FUjNZB7rK8HqKoGKlx21bBrM/KY3a4aovMaP4+vrEI/d9XQ2g2jproGHdxVw3X3A1HbyYP9OfbzS+5Y2FWjZvFydHJXDW15dmMvirtjRHFXDRO2M9k3TNl257Nz15T8YjuvQpDJgXSciRDaTx/azhu0nz60nTdoP31oO2/QfvoE6TgzVIMQQgghhBAJ6DgTQgghhBAiAR1nQgghhBBCJKDjTAghhBBCiAR0nAkhhBBCCJGAjjMhhBBCCCES0HEmhBBCCCFEAjrOhBBCCCGESEDHmRBCCCGEEAnoOBNCCCGEECIBHWdCCCGEEEIkoONMCCGEEEKIBHScCSGEEEIIkYCOMyGEEEIIIRLQcSaEEEIIIUQCOs6EEEIIIYRIELMsywpbCUIIIYQQQqLOqF5xfvzxx8NWoaih/fSh7bxB++lD23mD9tOHtvMG7adPkLYb1Y4zIYQQQgghfkHHmRBCCCGEEAkSa9euXRu2Eia54YYbwlahqKH99KHtvEH76UPbeYP204e28wbtp09QtmNyICGEEEIIIRIwVIMQQgghhBAJ6DgTQgghhBAiQUnYCvjNpUuX8OKLL6K1tRX19fVYuXIlJk2aZNu2t7cXK1euxJ133okf/OAHAWsaPWRs19raik2bNiGdTiOVSmHevHmYO3duSBpHBxnbffbZZ9ixYwf6+voQj8exYMECzJo1KySNo4XsvF2/fj0+/fRT3HzzzWN+66a//OUveO2115BOp/HNb34T3/72t3PODwwMYNu2bThx4gQqKiqwYsUKTJkyJSRto4eb/T766CPs2rULf//737FixQrMnDkzJE2jh5vtfvvb3+L3v/89EokEKisr8cADD6C+vj4kbaOHm/3279+Pt99+G/F4HGVlZfha1mSUAAAGLUlEQVTRj36Ea665JiRto4Wb7bL86U9/wubNm7FhwwZMmzbNXyWsUcbrr79u7d2717Isy9q7d6/1+uuvC9v+6le/sn7+859br7zySlDqRRoZ2w0MDFj9/f2WZVlWX1+f9eCDD1rt7e2B6hlFZGx3+vRp68yZM5ZlWVZ7e7v1wx/+0Lp06VKgekYV2Xl7+PBh69ChQ9aGDRuCVC9ypFIpa9myZVZLS4s1MDBgrV692jp16lROm7feesv65S9/aVmWZR08eNDavHlzGKpGEhn7nTt3zvrss8+srVu3Wu+//35ImkYPGdsdOXLESiaTlmVZ1ttvv82xNwIZ+/X09Az/fOjQIeu5554LWs1IImM7y7Ks3t5ea82aNdYTTzxhNTc3+67HqAvVOHToEGbPng0AmD17Ng4dOmTb7sSJE+jq6sJXvvKVINWLNDK2Kykpwbhx4wBkVrTS6XSgOkYVGds1NTXhqquuAgDU1NSgqqoKFy9eDFTPqCI7b2+99VZMmDAhSNUiSXNzMxobG9HQ0ICSkhLMmjWrwGZ//vOf8fWvfx0AMHPmTPztb3+DxVxwAHL2mzJlCr7whS8gFouFpGU0kbHdLbfcgtLSUgDAjTfeiI6OjjBUjSQy9isvLx/+OZlMcgwOIWM7AHjzzTfxrW99a9hX8ZtR5zh3dXWhuroaAFBdXW3rmKTTaezevRvf/e53g1Yv0sjYDgDa2tqwevVqPPDAA5g/fz5qamqCVDOSyNouS3NzMwYHB9HQ0BCEepFH1X5jnY6ODtTW1g7/XltbW+CcjGyTSCRQXl6O7u7uQPWMKjL2I/ao2u4Pf/gDbrvttiBUKwpk7ffWW29h+fLl2LNnD773ve8FqWJkkbHdyZMn0dbWhjvuuMOYHkUZ47xu3Tp0dnYWHF+4cKFU//379+OrX/0q6urq/FYt8ni1HQDU1dVh06ZN6OjowMaNGzFz5kxMnjzZTzUjiR+2A4ALFy5g69atWLp0KeLxUfe3qxC/7Edgu3Kcvyol02asQtvoo2K79957DydOnMAoLxehhKz95s2bh3nz5uHgwYP4zW9+g2XLlgWhXqRxs106ncauXbvw4IMPGtWjKB3np556SniuqqoKFy5cQHV1NS5cuIDKysqCNseOHcPHH3+M/fv3I5lMYnBwEGVlZVi0aJFJtSOBV9uNpKamBlOnTsUnn3wyJhJn/LBdb28vnn/+eSxcuBA33XSTKVUjiZ9jb6xTW1uL9vb24d/b29uHV+zz29TW1iKVSqG3t1eYKD3WkLEfsUfWdocPH8bevXuxdu1aY6/MixHVsTdr1izs2LEjCNUij5vtkskkTp06hWeeeQYA0NnZiZ/97Gd49NFHfU0QHHXLXdOnT8eBAwcAAAcOHMCMGTMK2vz4xz/GSy+9hO3bt+Pee+/FPffcMyacZjdkbNfe3o7+/n4AmZ0Qjh49iqampkD1jCIythscHMSmTZtwzz334Gtf+1rQKkYaGfuRK0ybNg1nz57F+fPnMTg4iD/+8Y+YPn16Tps77rgD7777LoBMhvmXv/xlrqoOIWM/Yo+M7U6ePIkdO3bg0UcfRVVVVUiaRhMZ+509e3b45w8//HA4N2as42a78vJyvPrqq9i+fTu2b9+OG2+80XenGRiFlQO7u7vx4osvoq2tDXV1dXj44YcxadIkHD9+HL/73e+wZMmSnPbvvvsujh8/zu3oIGe7w4cPY/fu3YjFYrAsC/PmzcOcOXPCVj10ZGz33nvv4aWXXsrZVmjp0qW47rrrwlM8IsjO2zVr1uD06dNIJpOoqKjAkiVLxmz85Icffohdu3YhnU7jG9/4BhYsWIA333wT06ZNw/Tp09Hf349t27bh5MmTmDRpElasWMGY+hG42a+5uRmbNm1CT08Pxo0bh8mTJ2Pz5s1hqx0J3Gy3bt06fP7558MhfHV1dXjsscdC1jo6uNnvtddew5EjR5BIJDBp0iR8//vfx9SpU8NWOxK42W4ka9euxb333kvHmRBCCCGEkDAYdaEahBBCCCGEmICOMyGEEEIIIRLQcSaEEEIIIUQCOs6EEEIIIYRIQMeZEEIIIYQQCeg4E0IIIYQQIgEdZ0IIIYQQQiT4T7S7bvM686B9AAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.scatter(y_hat, res)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Попробуем что-то из sklearn\n",
    "## Sentiment analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "В папке `/data/sentiment` лежат файлы, связанные с\n",
    "* imdb.com\n",
    "* amazon.com\n",
    "* yelp.com\n",
    "\n",
    "Формат файла:\n",
    "<Review>\\t<class label>\\n\n",
    "\n",
    "\n",
    "### Task\n",
    "1. Загрузите данные из файла\n",
    "2. Обучите линейный классификатор (для нас пока черный ящик)\n",
    "3. Попробуем проинтерпретировать результат\n",
    "4. Как будем оценикать качество?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "metadata": {},
   "outputs": [],
   "source": [
    "imdb = pd.read_csv('./data/sentiment/imdb_labelled.txt', sep='\\t', names=['text', 'label'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "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>text</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>A very, very, very slow-moving, aimless movie ...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Not sure who was more lost - the flat characte...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Attempting artiness with black &amp; white and cle...</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Very little music or anything to speak of.</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>The best scene in the movie was when Gerardo i...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                text  label\n",
       "0  A very, very, very slow-moving, aimless movie ...      0\n",
       "1  Not sure who was more lost - the flat characte...      0\n",
       "2  Attempting artiness with black & white and cle...      0\n",
       "3       Very little music or anything to speak of.        0\n",
       "4  The best scene in the movie was when Gerardo i...      1"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imdb.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import CountVectorizer"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [],
   "source": [
    "vect = CountVectorizer()\n",
    "X = vect.fit_transform(imdb.text.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(748, 2)"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "imdb.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<748x3047 sparse matrix of type '<class 'numpy.int64'>'\n",
       "\twith 11363 stored elements in Compressed Sparse Row format>"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LogisticRegression"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "model = LogisticRegression()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = imdb.label.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/andrey.shestakov/anaconda3/lib/python3.7/site-packages/sklearn/linear_model/logistic.py:432: FutureWarning: Default solver will be changed to 'lbfgs' in 0.22. Specify a solver to silence this warning.\n",
      "  FutureWarning)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
       "          intercept_scaling=1, max_iter=100, multi_class='warn',\n",
       "          n_jobs=None, penalty='l2', random_state=None, solver='warn',\n",
       "          tol=0.0001, verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "model.fit(X, y)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "coef = model.coef_[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "words = vect.get_feature_names()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "weights = pd.Series(index=words, data=coef)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "bad            -1.872306\n",
       "awful          -1.333721\n",
       "stupid         -1.175280\n",
       "no             -1.139851\n",
       "cheap          -1.139122\n",
       "even           -1.091030\n",
       "would          -1.050318\n",
       "worst          -1.038377\n",
       "waste          -1.037806\n",
       "nothing        -0.974179\n",
       "little         -0.969756\n",
       "boring         -0.883416\n",
       "just           -0.876937\n",
       "make           -0.873727\n",
       "how            -0.870910\n",
       "plot           -0.830202\n",
       "terrible       -0.787736\n",
       "annoying       -0.785370\n",
       "only           -0.777459\n",
       "ve             -0.770368\n",
       "mess           -0.738991\n",
       "shame          -0.721860\n",
       "free           -0.703715\n",
       "average        -0.696734\n",
       "can            -0.691914\n",
       "avoid          -0.685378\n",
       "directing      -0.680245\n",
       "half           -0.677415\n",
       "slow           -0.670796\n",
       "embarrassing   -0.655096\n",
       "                  ...   \n",
       "mostly          0.677508\n",
       "must            0.680133\n",
       "film            0.683956\n",
       "hilarious       0.692651\n",
       "give            0.693360\n",
       "long            0.693632\n",
       "still           0.694232\n",
       "nice            0.694942\n",
       "however         0.695025\n",
       "good            0.696534\n",
       "love            0.721136\n",
       "cool            0.745332\n",
       "entertaining    0.764841\n",
       "actually        0.789010\n",
       "best            0.798382\n",
       "played          0.820272\n",
       "cast            0.850574\n",
       "loved           0.893920\n",
       "interesting     0.894887\n",
       "enjoyed         0.925466\n",
       "right           0.986236\n",
       "excellent       1.008790\n",
       "brilliant       1.015302\n",
       "beautiful       1.043808\n",
       "well            1.044467\n",
       "10              1.054725\n",
       "great           1.069061\n",
       "funny           1.112703\n",
       "wonderful       1.242099\n",
       "liked           1.287456\n",
       "Length: 3047, dtype: float64"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "weights.sort_values()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "anaconda-cloud": {},
  "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": "3.7.0"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {
    "height": "142px",
    "width": "252px"
   },
   "number_sections": false,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {},
   "toc_section_display": "block",
   "toc_window_display": false
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}