homework_module_1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "6DZEWMGO16C9"
   },
   "source": [
    "# Домашнее задание. Титаник\n",
    "В этом домашнем задании вам предлагается решить задачу предсказания выживших пассажиров Титаника. Эту задачу машинного обучения рано или поздно решает, наверное, любой, кто обучается анализу данных. Информация о датасете доступна по <a href=https://www.kaggle.com/c/titanic>ссылке</a>.\n",
    "\n",
    "Кстати, здесь же есть много готовых решений нашей задачи. Решение можно посмотреть в разделе <a href=https://www.kaggle.com/c/titanic/code>code на kaggle</a>.\n",
    "\n",
    "Некоторые решения, которые уже есть:\n",
    "https://www.kaggle.com/hacktech33/titanic-solution-xgboost (простое решение)\n",
    "\n",
    "https://www.kaggle.com/shrutijhaa/in-top-3-titanic-machine-learning-from-disaster (возможно, слишком много визуализации, но хорошее качество)\n",
    "\n",
    "https://www.kaggle.com/blackhurt/top-3-using-voting-classifier (сильное решение, использует метод простого голосования трёх сильных моделей)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "F_a8jZs316DB"
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "8blX9KAN16DC"
   },
   "source": [
    "## Загрузка данных\n",
    "\n",
    "Загрузим датасет Titanic из приложенных файлов. В первом файле находится информация о пассажирах, во втором -- информация о том, выжил пассажир или нет.\n",
    "\n",
    "Если вы запускаете ноутбук на Google Colab, самым простым способом загрузить данные в ноутбук будет примонтировать к Colab ваш Google Drive. Для этого сначала положите в свою корневую директорию Google Drive  файлы, приложенные к этому ноутбуку, а затем выполните три закоментированные строчки в клетке ниже. <a href=https://www.machinelearningmastery.ru/downloading-datasets-into-google-drive-via-google-colab-bcb1b30b0166/>Подробная инструкция по работе с Google Drive + Google Colab</a>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "id": "2hL8eJ4A16DC"
   },
   "outputs": [],
   "source": [
    "#from google.colab import drive\n",
    "#drive.mount('/content/gdrive/')\n",
    "#data = pd.read_csv('/content/gdrive/My Drive/titanic_data.csv, index_col='PassengerId')\n",
    "\n",
    "#Если запускаете ноутбук локально:\n",
    "data = pd.read_csv('titanic_data.csv', index_col='PassengerId')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "id": "dIEKQ5g716DD",
    "outputId": "35f069fc-ee1f-4a66-d679-2ad00d1d521a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Всего 891 пассажиров в выборке\n"
     ]
    },
    {
     "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>Pclass</th>\n",
       "      <th>Name</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Ticket</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PassengerId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Pclass                                               Name  \\\n",
       "PassengerId                                                              \n",
       "1                 3                            Braund, Mr. Owen Harris   \n",
       "2                 1  Cumings, Mrs. John Bradley (Florence Briggs Th...   \n",
       "3                 3                             Heikkinen, Miss. Laina   \n",
       "4                 1       Futrelle, Mrs. Jacques Heath (Lily May Peel)   \n",
       "5                 3                           Allen, Mr. William Henry   \n",
       "\n",
       "                Sex   Age  SibSp  Parch            Ticket     Fare Cabin  \\\n",
       "PassengerId                                                                \n",
       "1              male  22.0      1      0         A/5 21171   7.2500   NaN   \n",
       "2            female  38.0      1      0          PC 17599  71.2833   C85   \n",
       "3            female  26.0      0      0  STON/O2. 3101282   7.9250   NaN   \n",
       "4            female  35.0      1      0            113803  53.1000  C123   \n",
       "5              male  35.0      0      0            373450   8.0500   NaN   \n",
       "\n",
       "            Embarked  \n",
       "PassengerId           \n",
       "1                  S  \n",
       "2                  C  \n",
       "3                  S  \n",
       "4                  S  \n",
       "5                  S  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "basic_features = data.columns\n",
    "\n",
    "y = pd.read_csv('titanic_surv.csv')\n",
    "y.index = data.index\n",
    "\n",
    "print(f'Всего {len(data)} пассажиров в выборке')\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "id": "cQDv_Jty16DE",
    "outputId": "022a59e4-ffbd-48dd-c10c-4f677b8ecf46"
   },
   "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>Survived</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PassengerId</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Survived\n",
       "PassengerId          \n",
       "1                   0\n",
       "2                   1\n",
       "3                   1\n",
       "4                   1\n",
       "5                   0"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "y.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "id": "R4_yv7cT16DE",
    "outputId": "87152d81-1afd-46ce-fd34-5c33e930e38d"
   },
   "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>Pclass</th>\n",
       "      <th>Name</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Ticket</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>Survived</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PassengerId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>NaN</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Pclass                                               Name  \\\n",
       "PassengerId                                                              \n",
       "1                 3                            Braund, Mr. Owen Harris   \n",
       "2                 1  Cumings, Mrs. John Bradley (Florence Briggs Th...   \n",
       "3                 3                             Heikkinen, Miss. Laina   \n",
       "4                 1       Futrelle, Mrs. Jacques Heath (Lily May Peel)   \n",
       "5                 3                           Allen, Mr. William Henry   \n",
       "\n",
       "                Sex   Age  SibSp  Parch            Ticket     Fare Cabin  \\\n",
       "PassengerId                                                                \n",
       "1              male  22.0      1      0         A/5 21171   7.2500   NaN   \n",
       "2            female  38.0      1      0          PC 17599  71.2833   C85   \n",
       "3            female  26.0      0      0  STON/O2. 3101282   7.9250   NaN   \n",
       "4            female  35.0      1      0            113803  53.1000  C123   \n",
       "5              male  35.0      0      0            373450   8.0500   NaN   \n",
       "\n",
       "            Embarked  Survived  \n",
       "PassengerId                     \n",
       "1                  S         0  \n",
       "2                  C         1  \n",
       "3                  S         1  \n",
       "4                  S         1  \n",
       "5                  S         0  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = data.join(y)\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "yrUiid0t16DF"
   },
   "source": [
    "## Исследование датасета\n",
    "\n",
    "### Задание 1 (1 балл)\n",
    "Опишите датасет. Сколько в нём мужчин, сколько женщин? Посчитайте распределение по классам пассажиров. Используйте функцию ``pd.Series.value_counts``. <a href=https://pandas.pydata.org/pandas-docs/stable/reference/api/pandas.Series.value_counts.html>Пример использования</a>\n",
    "\n",
    "Посчитайте долю выживших мужчин к общему количеству мужчин, а также женщин к общему количеству женщин. Сделайте выводы. \n",
    "Проведите схожую аналитику по возрасту выживших и по классу выживших. \n",
    "\n",
    "Вам поможет функция ``plt.hist()``. <a href=https://matplotlib.org/3.3.1/gallery/pyplots/pyplot_text.html#sphx-glr-gallery-pyplots-pyplot-text-py> Пример использования </a> "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "id": "t7jOomcB16DF"
   },
   "outputs": [
    {
     "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": [
    "survived_data = data[data['Survived']==1]\n",
    "plt.bar(['male', 'female'], [len(survived_data[survived_data['Sex']=='male'])/len(data[data['Sex']=='male']), len(survived_data[survived_data['Sex']=='female'])/len(data[data['Sex']=='female'])])\n",
    "plt.xlabel('Sex')\n",
    "plt.ylabel('Survival rate')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "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": [
    "survived_data = data[data['Survived']==1]\n",
    "plt.bar(range(1, 4), [len(survived_data[survived_data['Pclass']==i])/len(data[data['Pclass']==i]) for i in range(1, 4)])\n",
    "plt.xticks(range(1, 4))\n",
    "plt.xlabel('PClass')\n",
    "plt.ylabel('Survival rate')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "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": [
    "survived_data = data[data['Survived']==1]\n",
    "plt.bar([str(i) + '-' + str(i+9) for i in range(1, 81, 10)], [len(survived_data[survived_data['Age'].apply(lambda x: (x-1)//10)==i])/len(data[data['Age'].apply(lambda x: (x-1)//10)==i]) for i in range(8)])\n",
    "plt.xticks([str(i) + '-' + str(i+9) for i in range(1, 81, 10)])\n",
    "plt.xlabel('Age')\n",
    "plt.ylabel('Survival rate')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Первый график демонстрирует, что доля выживших женщин значительно превосходит долю выживших мужчин.\n",
    "\n",
    "Второй график демонстрирует, что процент выживших монотонно убывает с номером класса.\n",
    "\n",
    "Третий график демонстрирует, что самый большой процент выживших среди детей до 10 лет включительно. На втором месте пассажиры в возрасте 31-40 лет. Меньше всего доля выживших среди пассажиров старше 60."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "kess0Cip16DG"
   },
   "source": [
    "## Заполнение пропусков в данных\n",
    "В данных имеются пропуски в трёх колонках: "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {
    "id": "XbgGWK6z16DG",
    "outputId": "0e686905-877d-48fb-e3a7-e563220445f6"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Age', 'Cabin', 'Embarked']"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.columns[data.isna().any()].tolist()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "8zP_WCyK16DG"
   },
   "source": [
    "Для простоты заполним все пропуски в категориальных колонках новым классом \"0\", а в числовой колонке Age --- медианным значением. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "id": "-bLaKhN-16DH"
   },
   "outputs": [],
   "source": [
    "data.loc[:, ['Cabin', 'Embarked']] = data.loc[:, ['Cabin', 'Embarked']].fillna('0')\n",
    "data['Age'] = data['Age'].fillna(data['Age'].median())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "kolX5dfb16DH"
   },
   "source": [
    "## Подготовка плана тестирования\n",
    "\n",
    "### Задание 2. (0 баллов)\n",
    "Разделите выборку на обучающее и тестовое множество в соотношении 70:30."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "id": "d9bVGOL716DH"
   },
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "data_train, data_test = train_test_split(data, test_size=0.3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "xjMZ0GE_16DH"
   },
   "source": [
    "## Выделение новых признаков\n",
    "Теперь отложим тестовые данные и приступим к моделированию. Внимательно изучите данные. Можете ли вы выделить признаки, которые не указаны явно в таблице?\n",
    "\n",
    "### Задание 3 (1 балл)\n",
    " Сформируйте по крайней мере один новый признак и объясните ваш выбор. Пример признака сформирован за вас. \n",
    "\n",
    "Указания: \n",
    "- Пассажиров можно поделить на несколько классов по их именам.\n",
    "- Различных номеров кают слишком много. Но буквы в номерах кают указывают на их местоположение. (***Этот признак сформирован в примере***)\n",
    "- Возможно, имеет смысл отделить мальчиков от мужчин. Отделять девочек от женщин может быть не так важно\n",
    "- Другие идеи для признаков можно почерпнуть на форуме в обсуждении задачи на kaggle.com."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "id": "D_nbB_z-16DI",
    "outputId": "2c0d653e-cfa9-4af5-a47f-5f21b374a45e"
   },
   "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>Pclass</th>\n",
       "      <th>Name</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Ticket</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>Survived</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PassengerId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>0</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>0</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>0</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>887</th>\n",
       "      <td>2</td>\n",
       "      <td>Montvila, Rev. Juozas</td>\n",
       "      <td>male</td>\n",
       "      <td>27.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>211536</td>\n",
       "      <td>13.0000</td>\n",
       "      <td>0</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>888</th>\n",
       "      <td>1</td>\n",
       "      <td>Graham, Miss. Margaret Edith</td>\n",
       "      <td>female</td>\n",
       "      <td>19.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>112053</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>B42</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>889</th>\n",
       "      <td>3</td>\n",
       "      <td>Johnston, Miss. Catherine Helen \"Carrie\"</td>\n",
       "      <td>female</td>\n",
       "      <td>28.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>W./C. 6607</td>\n",
       "      <td>23.4500</td>\n",
       "      <td>0</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>890</th>\n",
       "      <td>1</td>\n",
       "      <td>Behr, Mr. Karl Howell</td>\n",
       "      <td>male</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>111369</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>C148</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>891</th>\n",
       "      <td>3</td>\n",
       "      <td>Dooley, Mr. Patrick</td>\n",
       "      <td>male</td>\n",
       "      <td>32.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>370376</td>\n",
       "      <td>7.7500</td>\n",
       "      <td>0</td>\n",
       "      <td>Q</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>891 rows × 11 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             Pclass                                               Name  \\\n",
       "PassengerId                                                              \n",
       "1                 3                            Braund, Mr. Owen Harris   \n",
       "2                 1  Cumings, Mrs. John Bradley (Florence Briggs Th...   \n",
       "3                 3                             Heikkinen, Miss. Laina   \n",
       "4                 1       Futrelle, Mrs. Jacques Heath (Lily May Peel)   \n",
       "5                 3                           Allen, Mr. William Henry   \n",
       "...             ...                                                ...   \n",
       "887               2                              Montvila, Rev. Juozas   \n",
       "888               1                       Graham, Miss. Margaret Edith   \n",
       "889               3           Johnston, Miss. Catherine Helen \"Carrie\"   \n",
       "890               1                              Behr, Mr. Karl Howell   \n",
       "891               3                                Dooley, Mr. Patrick   \n",
       "\n",
       "                Sex   Age  SibSp  Parch            Ticket     Fare Cabin  \\\n",
       "PassengerId                                                                \n",
       "1              male  22.0      1      0         A/5 21171   7.2500     0   \n",
       "2            female  38.0      1      0          PC 17599  71.2833   C85   \n",
       "3            female  26.0      0      0  STON/O2. 3101282   7.9250     0   \n",
       "4            female  35.0      1      0            113803  53.1000  C123   \n",
       "5              male  35.0      0      0            373450   8.0500     0   \n",
       "...             ...   ...    ...    ...               ...      ...   ...   \n",
       "887            male  27.0      0      0            211536  13.0000     0   \n",
       "888          female  19.0      0      0            112053  30.0000   B42   \n",
       "889          female  28.0      1      2        W./C. 6607  23.4500     0   \n",
       "890            male  26.0      0      0            111369  30.0000  C148   \n",
       "891            male  32.0      0      0            370376   7.7500     0   \n",
       "\n",
       "            Embarked  Survived  \n",
       "PassengerId                     \n",
       "1                  S         0  \n",
       "2                  C         1  \n",
       "3                  S         1  \n",
       "4                  S         1  \n",
       "5                  S         0  \n",
       "...              ...       ...  \n",
       "887                S         0  \n",
       "888                S         1  \n",
       "889                S         0  \n",
       "890                C         1  \n",
       "891                Q         0  \n",
       "\n",
       "[891 rows x 11 columns]"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "id": "HtBP15P916DI",
    "outputId": "d2f2b24b-7bc5-4697-bec7-b881a05056a9"
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "PassengerId\n",
       "1         0\n",
       "2       C85\n",
       "3         0\n",
       "4      C123\n",
       "5         0\n",
       "       ... \n",
       "887       0\n",
       "888     B42\n",
       "889       0\n",
       "890    C148\n",
       "891       0\n",
       "Name: Cabin, Length: 891, dtype: object"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['Cabin']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "id": "HZAaYeAe16DI"
   },
   "outputs": [],
   "source": [
    "def get_cabin_letter(row):\n",
    "    return row['Cabin'][0]\n",
    "\n",
    "data['cabin_type'] = data.apply(get_cabin_letter, axis=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "XPWVtLul16DJ"
   },
   "source": [
    "Гонорифики -- слова стоящие перед именем в уважительном обращении. По ним можно определить примерную социальную группу пассажира. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "id": "11qopJCo16DJ",
    "outputId": "f6921fcd-d223-4e84-921c-c586279211be"
   },
   "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>Pclass</th>\n",
       "      <th>Name</th>\n",
       "      <th>Sex</th>\n",
       "      <th>Age</th>\n",
       "      <th>SibSp</th>\n",
       "      <th>Parch</th>\n",
       "      <th>Ticket</th>\n",
       "      <th>Fare</th>\n",
       "      <th>Cabin</th>\n",
       "      <th>Embarked</th>\n",
       "      <th>Survived</th>\n",
       "      <th>cabin_type</th>\n",
       "      <th>honorific</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PassengerId</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3</td>\n",
       "      <td>Braund, Mr. Owen Harris</td>\n",
       "      <td>male</td>\n",
       "      <td>22.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>A/5 21171</td>\n",
       "      <td>7.2500</td>\n",
       "      <td>0</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Mr.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>Cumings, Mrs. John Bradley (Florence Briggs Th...</td>\n",
       "      <td>female</td>\n",
       "      <td>38.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>PC 17599</td>\n",
       "      <td>71.2833</td>\n",
       "      <td>C85</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "      <td>C</td>\n",
       "      <td>Mrs.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>Heikkinen, Miss. Laina</td>\n",
       "      <td>female</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>STON/O2. 3101282</td>\n",
       "      <td>7.9250</td>\n",
       "      <td>0</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>Miss.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>Futrelle, Mrs. Jacques Heath (Lily May Peel)</td>\n",
       "      <td>female</td>\n",
       "      <td>35.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>113803</td>\n",
       "      <td>53.1000</td>\n",
       "      <td>C123</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "      <td>C</td>\n",
       "      <td>Mrs.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>3</td>\n",
       "      <td>Allen, Mr. William Henry</td>\n",
       "      <td>male</td>\n",
       "      <td>35.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>373450</td>\n",
       "      <td>8.0500</td>\n",
       "      <td>0</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Mr.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>887</th>\n",
       "      <td>2</td>\n",
       "      <td>Montvila, Rev. Juozas</td>\n",
       "      <td>male</td>\n",
       "      <td>27.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>211536</td>\n",
       "      <td>13.0000</td>\n",
       "      <td>0</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Rev.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>888</th>\n",
       "      <td>1</td>\n",
       "      <td>Graham, Miss. Margaret Edith</td>\n",
       "      <td>female</td>\n",
       "      <td>19.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>112053</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>B42</td>\n",
       "      <td>S</td>\n",
       "      <td>1</td>\n",
       "      <td>B</td>\n",
       "      <td>Miss.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>889</th>\n",
       "      <td>3</td>\n",
       "      <td>Johnston, Miss. Catherine Helen \"Carrie\"</td>\n",
       "      <td>female</td>\n",
       "      <td>28.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>W./C. 6607</td>\n",
       "      <td>23.4500</td>\n",
       "      <td>0</td>\n",
       "      <td>S</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Miss.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>890</th>\n",
       "      <td>1</td>\n",
       "      <td>Behr, Mr. Karl Howell</td>\n",
       "      <td>male</td>\n",
       "      <td>26.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>111369</td>\n",
       "      <td>30.0000</td>\n",
       "      <td>C148</td>\n",
       "      <td>C</td>\n",
       "      <td>1</td>\n",
       "      <td>C</td>\n",
       "      <td>Mr.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>891</th>\n",
       "      <td>3</td>\n",
       "      <td>Dooley, Mr. Patrick</td>\n",
       "      <td>male</td>\n",
       "      <td>32.0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>370376</td>\n",
       "      <td>7.7500</td>\n",
       "      <td>0</td>\n",
       "      <td>Q</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>Mr.</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>891 rows × 13 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "             Pclass                                               Name  \\\n",
       "PassengerId                                                              \n",
       "1                 3                            Braund, Mr. Owen Harris   \n",
       "2                 1  Cumings, Mrs. John Bradley (Florence Briggs Th...   \n",
       "3                 3                             Heikkinen, Miss. Laina   \n",
       "4                 1       Futrelle, Mrs. Jacques Heath (Lily May Peel)   \n",
       "5                 3                           Allen, Mr. William Henry   \n",
       "...             ...                                                ...   \n",
       "887               2                              Montvila, Rev. Juozas   \n",
       "888               1                       Graham, Miss. Margaret Edith   \n",
       "889               3           Johnston, Miss. Catherine Helen \"Carrie\"   \n",
       "890               1                              Behr, Mr. Karl Howell   \n",
       "891               3                                Dooley, Mr. Patrick   \n",
       "\n",
       "                Sex   Age  SibSp  Parch            Ticket     Fare Cabin  \\\n",
       "PassengerId                                                                \n",
       "1              male  22.0      1      0         A/5 21171   7.2500     0   \n",
       "2            female  38.0      1      0          PC 17599  71.2833   C85   \n",
       "3            female  26.0      0      0  STON/O2. 3101282   7.9250     0   \n",
       "4            female  35.0      1      0            113803  53.1000  C123   \n",
       "5              male  35.0      0      0            373450   8.0500     0   \n",
       "...             ...   ...    ...    ...               ...      ...   ...   \n",
       "887            male  27.0      0      0            211536  13.0000     0   \n",
       "888          female  19.0      0      0            112053  30.0000   B42   \n",
       "889          female  28.0      1      2        W./C. 6607  23.4500     0   \n",
       "890            male  26.0      0      0            111369  30.0000  C148   \n",
       "891            male  32.0      0      0            370376   7.7500     0   \n",
       "\n",
       "            Embarked  Survived cabin_type honorific  \n",
       "PassengerId                                          \n",
       "1                  S         0          0       Mr.  \n",
       "2                  C         1          C      Mrs.  \n",
       "3                  S         1          0     Miss.  \n",
       "4                  S         1          C      Mrs.  \n",
       "5                  S         0          0       Mr.  \n",
       "...              ...       ...        ...       ...  \n",
       "887                S         0          0      Rev.  \n",
       "888                S         1          B     Miss.  \n",
       "889                S         0          0     Miss.  \n",
       "890                C         1          C       Mr.  \n",
       "891                Q         0          0       Mr.  \n",
       "\n",
       "[891 rows x 13 columns]"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def get_honorific(row): \n",
    "    return row['Name'][row['Name'].find(',') + 2: row['Name'].find('.')+1]\n",
    "\n",
    "data['honorific'] = data.apply(get_honorific, axis=1) \n",
    "data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "_zbESUzf16DJ"
   },
   "source": [
    "## Кодирование категориальных признаков и обработка данных\n",
    "Все признаки сейчас делятся на числовые, бинарные, категориальные и текстовые. К текстовым признакам можно отнести имя пассажира, номер билета и номер каюты: значения этих признаков уникальны почти для всех пассажиров, и простого способа использовать их в модели не существует.\n",
    "\n",
    "### Задание 4 (0 баллов)\n",
    "Пока используя только изначальные признаки, перекодируйте категориальные признаки функцией ``pd.get_dummies``. Естественно, имена пассажиров и номера билетов и кают кодировать не стоит, поскольку все значения этих признаков уникальные. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "id": "GXNL-Ak016DJ"
   },
   "outputs": [],
   "source": [
    "X_train = pd.get_dummies(data_train, columns=['Sex','Embarked']).drop(columns=['Name', 'Ticket', 'Cabin', 'Survived'])\n",
    "y_train = data_train['Survived']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "YLtzDDYE16DJ"
   },
   "source": [
    "## Обучение baseline-модели\n",
    "\n",
    "### Задание 5 (1 балл)\n",
    "\n",
    "Обучите одну из простых известных вам моделей. Измерьте качество полученной модели на кросс-валидации (используйте только обучающую выборку!)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "id": "PTT0sR5w16DK"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [00:02<00:00, 36.04it/s]\n"
     ]
    },
    {
     "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": [
    "from sklearn.neighbors import KNeighborsClassifier\n",
    "from sklearn.model_selection import cross_validate\n",
    "from tqdm import tqdm\n",
    "\n",
    "scores = []\n",
    "for n in tqdm(range(1, 101)):\n",
    "    model = KNeighborsClassifier(n_neighbors = n)\n",
    "    scores.append(cross_validate(model, X_train, y_train)['test_score'].mean())\n",
    "plt.plot(scores)\n",
    "plt.xlabel('n_neighbors')\n",
    "plt.ylabel('cross-validation score')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "PTHGroZo16DK"
   },
   "source": [
    "### Задание 6 (1 балл)\n",
    "Добавьте в модель два новых сгенерированных признака. Если нужно, добавьте признаки, сгенерированные ohe-hot кодированием. Обучите ту же самую модель на расширенном множестве признаков. Улучшилось ли качество предсказания на кросс-валидации?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {
    "id": "fQTcuXHV16DK"
   },
   "outputs": [],
   "source": [
    "X = pd.get_dummies(data, columns=['Sex','Embarked', 'cabin_type', 'honorific']).drop(columns=['Name', 'Ticket', 'Cabin', 'Survived'])\n",
    "y = data['Survived']\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [00:03<00:00, 32.68it/s]\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYgAAAEHCAYAAAC0pdErAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAABLfklEQVR4nO29e3xc9Xnn/37mrttIlmxLtiVLNrEBA7YBx+DcSAhJITdyXyCENO02pU2aptvNNtluNr1st5vNtr/tJQklaRIgF3IjgWQpkIZbGzBgwDIGbGNAsmRbtiV7dJ/79/fHOWc0M5qRzmg0kjx63q/XvKQ5c86c7xHmfM5zF2MMiqIoipKPZ7EXoCiKoixNVCAURVGUgqhAKIqiKAVRgVAURVEKogKhKIqiFMS32AuYT1auXGm6uroWexmKoihnDU8//fSgMWZVoc+qSiC6urrYs2fPYi9DURTlrEFEeot9pi4mRVEUpSAqEIqiKEpBVCAURVGUgqhAKIqiKAVRgVAURVEKogKhKIqiFEQFQlEURSmICkSZnByJsq8/stjLUBRFmXdUIMrkqw+/zG99W4vzFEWpPlQgymQ8lmRoPEYqrYOXFEWpLlQgyiSRSmMMjEYTi70URVGUeUUFokziqTQAkQkVCEVRqgsViDKJJy3XUmRSBUJRlOpCBaJMEhkLIr7IK1EURZlfVCDKJJ60BGJYLQhFUaoMFYgySZQYgzgxEuU/3vYUZ8bV4lAUZWmjAlEmpQrEs0fO8K8vnuTRl05VclmKoihlowJRJjHbxRSZdGcRjMVSADx7JFKpJSmKoswLKhBl4lgQwy4tiDG7XmJvX6RSS1IURZkXVCDKJJEqLc11PG5ZEC8cGyGWTFVsXYqiKOWiAlEmThaT2zTX0WjSOi6V5sXjoxVbl6IoSrmoQJRJJkjt1oKIJfF5BIC9R85UbF2KoijlogJRJsVabRwcGOULP9tPOq+J31gsyZqmEKsbghqHUBRlSaMCUSbZLqZsMbj3uePcsbuXwfFYzv5jsST1QT/bO5pUIBRFWdKoQJRJIpUm5PeQNjAWT2a2nxiJAjAymczZfyyapD7oZfv6JnqGJrRgTlGUJUtFBUJErhaRgyJyWEQ+V+Dzz4rIXvu1X0RSItJsf/ZNETkpIvsrucZySKUNaQOrGoJAbqrrgC0Q+W3Ax+NJ6oM+tnc0AbBXp9EpirJEqZhAiIgX+ApwDbAFuF5EtmTvY4z5sjFmuzFmO/B54BFjzGn7428DV1dqffOB415a3RACcuMQA8OOQEy3IOqCPra2NyECe7VgTlGUJUolLYidwGFjzCvGmDhwJ3DtDPtfD3zfeWOMeRQ4XXz3xccJUK+qtyyI7Gpqx4IYybMgxmJJGkI+6oM+Nq9u0DiEoihLlkoKxDqgL+t9v71tGiJSi2Ut/KTUk4jIJ0Rkj4jsOXVqYfsbOSmujovJsSCiiVTm92kWRCxJXcAHwPaOJrr7Ixij40oVRVl6VFIgpMC2YnfCdwO/znIvucYYc6sxZocxZseqVatKPbwsplxMjgVhiYIToIbcGEQqbZiIp6gPWQJx8fomIhMJeoYmFmrJiqIorqmkQPQDHVnv24FjRfa9jiz30tmCY0GszASpLReTE3+A3CymcTvLqT5oCcQ2O1DdrW4mRVGWIJUUiKeATSKyQUQCWCJwT/5OItIIXAHcXcG1VARHIOqCPmoD3oxbaaCIBTEeyxWITavrCfk9dGsmk6IoS5CKCYQxJgl8CrgfeBH4oTHmeRG5WURuztr1fcADxpjx7ONF5PvA48C5ItIvIr9dqbXOFafVd8DrYUVtIONiciyIplp/TgxizP69zhYIn9fDhWsb2dc/vJDLVhRFcYWvkl9ujLkXuDdv2y1577+NldKaf+z1lVzbfOB0cg34hMYaf44FURfwsraxJieLacyxIEJTf/ZtHU1894lekqk0Pq/WLSqKsnTQO1IZOC4mv9dDU60/09F1YDhKa2OIhpCPkWwLIs/FBLC1vZFoIs2hE2MLuHJFUZTZUYEog3iWi6mp1j/lYhqJsqYxREMo18WUH4MA2NbeBMA+jUMoirLEUIEoA6dQzu/z0FgTyLiYTgxHaQ2HCNf4GMlqA+6IRbZAdLbU0ljjp3sR4xDjsSQPPD+waOdXFGVpogJRBok8C2J4Mk4qbTg5GqMtHCIc8s+YxQQgImxtb1zUVNe/f/AlPnHH0xwfnly0NSiKsvRQgSgDx4II+Dw01fhJpAxHTk+QTBva7BjEWCyZaQPuxCDqgrm5AVvbGzl4YpRoYuFHkEYTKX74lFXwfnIkNsveiqIsJ1QgyiA/SA1wcGAEIGNBpM1UgdxYLEXA5yHgy/2zb21vIpU2PH9sZAFXb/H/9h3njO0aO62txxVFyUIFogwSScsy8HuFxpoAAAcGrDnTjgUBU7GHsVgix73ksJiB6jt292bEbXBMLQhFUaZQgSiDHBdTxoKwBSJsZTHBlECMx1IFBaKt0RpButAFc8/1D7O3L8LvvHEjoBaEoii5qECUQX6aK1gWhM8jtNQHCddYYuAUy43asyAKsbW9acFbbnxndy81fi83Xt5J0OdhSAVCUZQsVCDKICcGYbuYeobGWd0QxOuRLAvCEojxmDVutBDb2ht55dT4tPkRlWJ4IsHd3Ud578Vraazxs7I+yNCYCoSiKFOoQJRBooCLyRhobbQmzE2PQSQLupgAttqdXffP4ma67tbH+fqjr5S99nu6jxJNpLnx8k4AmusCnB7XGISiKFOoQJSB42LyeYSQ30vIb/0519gCEbYtCKdYbjxW3MX0mtX1APSdKT4bwhjDM70RHnih/KK2p3rOsKYxxAVrGwFoqQ+oi0lRlBxUIMognjIEvB5ErNlIjpupNZxrQTj9mEbtcaOFyLc2CjGZSBFPpdl/dISkbb3MlX39Eba2N2beN9cF1MWkKEoOKhBlkEilc2oaHDdTmy0QIb+XgNeTlcU0NW40n/qADxFyWnPk47TymEykeOnk3Jv7DdtT7Lba6bWAFYNQF5OiKFmoQJRBPJnG752arNpYYwuE7WICrH5M0cS0caP5eDxCfTC3+2s+jkBAeTUT+45ax27LEojmugDRRJqJePHzK4qyvFCBKINEKo3fW9yCADIdXfPHjRYiHPLPmMUUmZxyAZXT3M+pt7goz8UEqJtJUZQMKhBlEM93MdkxiGwLoiHkYzSayEyTm0kgrH2LP8E77qeWukBZFkR3X4QNK+syFg/AynpbIDRQrSiKjQpEGcSTaQLZFkSddcNtzbIgwiE/I5OJTCfXYllMzr6jM1kQtovpjZtWcuD43Jv77esfzglQAzTXBQEY0nYbiqLYqECUQb6L6T/s6OCv338RIf9UMZxjFYwWGDeajzU/YoYYhG1BXHHuKpJpwwvHS2/ud3IkysBINCdADZZVAmpBKIoyRUVnUlc7iZTJcTFtXFXPxlX1OftYY0cTBWdB5NMQ8jMaGy36eWQigd8rXL6xBYB9fREuWb+ipDU7sYtteRZES73GIBRFyUUtiDLIz2IqRNgOUruJQYRDM1sQw5NxGmsCtIVDrJpjc799/RG8HskUyDnUBnzU+L1aTa0oSoZZBUJENovIr0Rkv/1+q4j8t8ovbekTz3MxFaIh5GcinmLYdg/NZkGMxZIYYwp+HplI0FTrR0TY1t44p+Z+3f3DbFpdT01gek8oLZZTFCUbNxbE14HPAwkAY8w+4LpKLupsIb9QrhBOR9fjw1Fg9iwmp16iEMOTCZrszKOt7U28XGJzP2MM+/ojOfUP2azUdhuKomThRiBqjTFP5m3TaiqmZzEVwuno6sx7njGLyb75F7vpOxYEkMlCmq25XzZ9pyeJTCTY2tFY8PPmuoBWUyuKksGNQAyKyDmAARCRDwLHK7qqs4T8LKZCOD2Wjg9HC44bLbRvsVqI4clEZnKdYwXMVjAXmYjz0MGTPP7yEL947ljOsfm01Ac5XcTFlE4b7ni8RyutFWUZ4SaL6ZPArcB5InIUeBX4SEVXdZaQSBn8s7mYbAviWGRyRvcSMG1+RD6RiXimuG1FXYCO5hr2H5tZIL503wG+/2Rf5n1twMu5bQ0F922pCzA4HscYk2lA6LC3P8IX7n6emoCPD17aPuM5FUWpDma8Y4mIF/g9Y8xVIlIHeIwxxfMwlxnuXExTFsTK+uCM+4ad7q8FMpkSqTTj8VTGxQSwcWU9PYPjM37nkdMTnNfWwBfffQHRRIq2xlBRq6e5LkA8aZ0nX8x6h8ZzfiqKUv3MKBDGmJSIXGr/rneGPKxWG7OnuQJMxFMzxh9gyoIoFINwsqCyBaKrpZane88UfOJ3GBiOsrm1gV3ntMx4brBcTACnx+LTBKJn0JpT0TNUfF6FoijVhZsYxLMico+IfFRE3u+8Kr6yswA3MQgniwmgYRaBmJphPd2CcNpsZPdP6mypYyyW5PQMmUcnR2I5rT9mwqmmHiwQqFYLQlGWH25iEM3AEHBl1jYD3FWRFZ1FuHExZT+J1xWZR+0QniEGMWx3cm2qDWS2da2sBayn+pYC7qvxmNXiw7VA2NXUhQLVvadtC2IWl5aiKNXDrAJhjPn4QizkbCSRSs8apPZ5PdQGvPYsCP+M+wZ9HvxeKZjF5FgQTXkWBFhP9Zd2Tm+5MTBi1V60Nc4c+3DItPwuaEFM4PUII9EkkYl4jlApilKduKmkbheRn4rISRE5ISI/EZFln8ZijLGymGaxIGDKMqifxYIQkUz313wKuZjaV9TgkeJxgRO2QLh3MdkdXfNcViPRBKfH41yyvgnQOISiLBfcxCC+BdwDrAXWAT+3t1UlP366n6d7z8y6X9yeCR2cxYKAqUym2dJcnX0LWhAFgtRBn5e1TTVF4wKlCkRNwEttwDut3cYRWxDetGkVoHEIRVkuuBGIVcaYbxljkvbr28CqCq9r0fjy/Qf47u7eWfdLpKx+SbM164MpgZgti8nat/BUueHJBCJTmU4OXS11RZ/oB4YtV1GbS4EAKw6RH/TusQXhDZtWApa7SVGU6sdtJfWNIuK1XzdiBa2rklgyXbQXUjaJpGVBuHIx1TguptkFIlxT2IIYnogTDvnxenIFaX1L7YwWREPQ50qYHJrrggzmDQ1yBOHctgbWNIYygqEoSnXjRiB+C/gwMIDVYuOD9raqJJ5MM+FiUpvjYpqtWR9MPfW7cjEFC0+Vi0wmctxLDl0ttUQmEgxPTD/mxEiU1WF3AWqHlXXTLYjeoXFWNQSpDfjobKlVC0JRlgmz3t2MMUeMMe8xxqwyxqw2xrzXGDO7DwYQkatF5KCIHBaRzxX4/LMistd+7ReRlIg0uzm2UsSTaSZd9BuKl2BBZGIQM0yTcyg2VS4ykcjJYHLIZDKdnv5UPzASzZmP7YZCLb97hiboarFSarta6jQGoSjLBDdZTLeJSFPW+xUi8k0Xx3mBrwDXAFuA60VkS/Y+xpgvG2O2G2O2Y7UUf8QYc9rNsZUglTYk04ZJFxZEwrEgSshichuDKGZBhAsIRJctEIXiEKUUyTk02zGI7JkUR4YmMkLU2VLH4FicsZg27VOUaseNi2mrMSbivDHGnAEudnHcTuCwMeYVY0wcuBO4dob9rwe+P8dj5wXHKnATgyjNxWQJw2yV1M6+4/EUSfv7HYaL1B6sb7ae7HvzCtjSacOJkWjJArGyLkg8lc7M0J6MpxgYidJpn6fTtiTUilCU6seNQHhEJFOFZbuA3EQ91wF9We/77W3TEJFa4GrgJ6UeO584AjHpKkjtZDG5sSDcZzE51kb+E3r2sKBsagJe2sKhaRbE0HicZNqUlMEEU8Vyg6NWoPqIXUHdudKxIByBmG6xpNOGm+94ml/sO1bSORVFWZq4EYi/AR4Tkb8Ukb8EHgP+t4vjCuV/Fp6lCe8Gfm2MOV3qsSLyCRHZIyJ7Tp065WJZxYmlLGFw42JyLAg3aa5b1oZpDQdZ21Qz676FZkKk08YSiAJBasAOHOc+0ZdaA+FwkT2I6JcvnACmLAUnBtGZcWlNtyAefekU9z0/wE+fOVrSORVFWZq4CVLfDnwAOAGcBN5vjLnDxXf3Ax1Z79uBYo+W1zHlXirpWGPMrcaYHcaYHatWlVeeUYqLKVGCi+nSzmae+K9X5VRBF8PJeBrOqqYejSVJG4oeX6gWYkogSsti2tzawM4NzXzniV5SaZOxFDqbLWGoD/pYWR/MFM9lc9tjPQB090eKztVWFOXswU2Q+hzgZWPMPwLPAVdlB61n4Clgk4hsEJEAlgjcU+D7G4ErgLtLPXa+cQQinkyTSs98g3P2dROkLgWno2u2BeGksBbrf7S+pZbBsRjjWW6pqT5MpVkQAB/b1UXf6UkePniSnqFxmmr9NGZZL50ttdMsiJ7BcR4+dIr2FTUMjsU5Gpks+byKoiwt3NzdfgKkROQ1wDeADcD3ZjvIGJMEPgXcD7wI/NAY87yI3CwiN2ft+j7ggex5E8WOdXlNcyaWnAoMzzZaM5Fyn+ZaCuECMyEidifXmSwIyI0LnBiJIQKrZhlSVIi3X9BKazjIbY/3cuT0RCZA7VCoFuKO3b14Rfizd18AQHef+1nZiqIsTdzc3dL2Dfv9wN8ZY/4IWOPmy40x9xpjNhtjzjHG/JW97RZjzC1Z+3zbGHOdm2MrTTxLIGaLQ5TiYiqFqZbfUwKV6eQ6QwwCcjOLTtgT7HxzEDC/18NHLuvk0UOn2Hskkok7OHS11HF8OErU/huNx5L8cE8f11y0hjdtXkXA56G7P1LyeRVFWVq4uXskROR64CbgF/a22Z3pZyHxrNTS2TKZYiUUypXCVJB6yoLITJMrYkE4ApEdhxgYiZacwZTNdTs7rNbjsWQmQJ1/vj47w+mnzx5lNJrkY7s6Cfg8XLA2zN4jkTmfW1GUpYGbu9vHgV3AXxljXhWRDcB3KrusxSGe42LKFQhjDOmsuITTrG++YxD1BeZSO51cG4tYEA0hPyvrA7kWxBxqILJZ3RDimgstQ3F9ngXhWBR/88Ah/tvPnuOrDx3mgrXhzEyKbe1NPHd0eFoth6IoZxduspheMMZ82hjzffv9q8aY/1X5pS08M7mY/vOP9vHpO5/NvK+Ui8lvDxjKsSAmZo5BAGxcVc/evqnsIUsgSo8/ZPM7b9xIQ8jH9o6mnO2bVteztjHEoy+d4l+eG8DjET5z1ebMXOztHU1MJlK8dHKsrPMrirK4uG/zuQyIJadEId/FdPjUGLEs0ZjqxTR7HUSpNIR8uUHqiQS1AS9BX/GBQ+/etpYv/Gw/z/ZF2LImzJmJRFkuJrBqIvZ98e2ZG79DXdDHY59/a9HjttmC0t0X4fw14bLWoCjK4jG/j79nObEZXEzjsWRO4DiTxTTPFgQ4/ZhyXUyz1VC87+J11AW8fGd3L6fsKujWOaS45pMvDm7oaqklHPJpoFpRznJUILKIz5DmOh5L5jzVx0to1lcq4bypcsMuBKI+6ON9l6zjF/uOc2BgFCi9inq+EBG2dTSxV1NdFeWsxk2h3GYR+bqIPCAiDzqvhVjcQpOdxRTNi0GMRZOMxZKZQHUp7b5LJX+q3PBE8TYb2dx4eSfxZJp/fOgwUNokuflme0cTh06MzlpPoijK0sVNDOJHwC3A14HZe1CcxRTLYjLGMBZPYgyMxZOEQ34SqTRej0yb8DYfhGv8mSZ5YBXKbVxZP+tx57WF2dG5gj32TO3FFIht7U2k0ob9R0fYuaF50dahKMrccfP4mzTGfM0Y86Qx5mnnVfGVLQLFBGIinsJpLTRip5wmUqYi7iWwgtSjeUFqNxYEwEd3dQIQ9HkybTsWg60dVtO/7r7Ioq1BUZTycHOH+7mI/L6IrBGRZudV8ZUtAtlB6mwXU3aPIyc2EE+mK5LBBHYWk10HkUyliUzMHoNwuPrCNlrqArQ1huYUYJ4vVjeEWN9cy6MvlddhV1GUxcPNI+bH7J+fzdpmgI3zv5zFJZ5M4xGo8XtzLIixQgKRSs97DYRDOOQnnkoTTaR45NAp4qk0F69fMfuBQNDn5c+vvYCx6OL7/t938Tr+/sGXODI0wfq8amxFUZY+bgrlNhR4VZ04wNRNvybgKyoQGRdTMl0xF5MzYGgkmuA7u3tZ0xjiqvNXuz7+XVvXct3O9RVZWylct7MDAb7/1JHFXoqiKHPATRaTX0Q+LSI/tl+fEpHq7MVk3/RrAh4ms7JvciyImCUQ8VS6IjUQMDUTYl/fMP/20iA37Fw/p6Z7i82axhquPK+VH+3py4nvKIpyduDmrvM14FLgq/brUntb1RFLpgn4vNT6fTmtNsZjU787sYFEKl2RFFeYmgnx1YcP4/fKkrAG5spHLl/P4FicB14YWOylKIpSIm5iEK81xmzLev+giHRXakGLSTyZJujzUBPIj0FkTXezs4viyUpmMVkWxDNHIrxn21pWNZTXU2kxedOmVbSvqOF7TxzhXVvXLvZyFEUpATd3uJQ9VQ4AEdlIldZDxJIpgj6rWd5kjkBM/Z4dpK6Ui8mZCQFwk522erbi9QjX71zPYy8P8copbd6nKGcTbu5wnwUeEpGHReQR4EHgjyu7rMUhnrSD1H5vnovJEoX64FQTPStIXbk0V4Dz10y10D6b+dCOdnwe4c6n+qZ9dk/3Mf7xwZcWYVWKoszGrC4mY8yvRGQTcC4gwAFjTKziK1sEprKY8iyIaBKPwOqGICPRqRhE0F8ZC6KlPsDGlXX8wZWvWdRahvlidUOIi9c3sbdA0dyPn+7nwPERPnXlpoVfmKIoM1JUIETkSmPMgyLy/ryPzhERjDF3VXhtC46TxVQ7LQaRpC7go6HGn+Nicob7zDdBn5cH//ObK/Ldi0VbYw37CnR3PTEc5fR4HGNMVYiholQTM93hrsByJ727wGcGqE6BsF1M2U3mxmNJ6kM+wiFfpg7CqqQ++1JPF4u2cJAHhqPThODEaJRk2jAymSw6MU9RlMWhqEAYY75o//oXxphXsz+zx45WHY5VUBPwEU1M5e2PxZLUBX2EQ36ORSYBy8VUqUrqaqQ1HCKWTDM8maCpNgBY7UwiE5bgDo7HVCAUZYnh5g73kwLbfjzfC1kKxBLpTBZTPJXOzFTOCESNLxODiKcqV0ldjbTZw4sGRqKZbSeyfj89Hl/wNSmKMjMzxSDOAy4AGvPiEGFg8fpIVxArSO2lNmCN9pxIpAh7PYzFkjQEffakNyeLyVSsWV814rQeHxiOcl5bOPO7w9BYVeY9KMpZzUwxiHOBdwFN5MYhRoHfqeCaFg0nSB3yWwIRjacIh/yMx5KsbgjSELRcT/FkWl1MJeJMt8u2GrKtiaEyLQhjDI+/MsTlG1rwVGBGh6IsR2aKQdwN3C0iu4wxjy/gmhaNmB2kzlgQdibTeCxFfdBP2G65PRpNaJC6RFozFsSUpZAtFkNj5QnE3r4IN3z9Cb5x0w6u2tJa1ncpimLhJk/zWRH5JJa7KeNaMsb8VsVWtUjEsyqpYUogRqMJ6oPeTAHbaDSpMYgSCfg8tNQFcqyGgeEYtQEvXpGyXUyHT1pV2odPjXEVKhCKMh+4ucPdAbQBvwE8ArRjuZmqDseCcFxMk4kkxhjG46lMFhNYbbjVxVQ6reFQjtVwYiRKWzjEyoZg2S6m3qEJ++d4Wd+jKMoUbu5wrzHGfAEYN8bcBrwTuKiyy1p4jDHEU04Wk2UpTMbTxJJpUmlDfciXsSAiEwnSBnUxlUhbYygnMH1iJEprOERzXaBsF1OPLQw9gxOz7Kkoilvc3OGcVqYREbkQaAS6KraiRSKZNhhDppIaYCKezFRO19tZTABD45Y7RAWiNFrDwWlB6rbGEC11gbLTXNWCUJT5x80d7lYRWQF8AbgHeAH43xVd1SLgDLRxejEBTCZSmUZ9dQFfZk6D87SrLqbSaA2HGBqPE0umMMZwciTG6nCQlvpARnTngjGGnqFxRODYcDRnnriiKHPHTbO+b9i/PkIVzqF2yBEIJwYRT2WmyVkuJseCsAVC6yBKwqmFODkSyxQjtoVD+D0eTo/HSafNnFJUz0wkGI0m2dbRRHdfhL7TE2xqbZjv5SvKsmOmQrn/NNOBxpi/nf/lLB7x1JRAZGcxjWW1+m4I+hCZKupSF1NptNrV1CdHo5lEAEc00gYikwma6wIlf++rg5Zb6YrNq+jui9AzpAKhKPPBTHe4Bvu1A/g9YJ39uhnYUvmlLSwxu/dS0Oct6GKqD/rweIT6gE9dTHOkLasWwolFtDaGMqIw11RXJ+5wxeZVOe8VRSmPmQrl/hxARB4ALjHGjNrv/wz40YKsbgGJpyy/dcDnIeD14BErSO1YEHVB60/VEPJlXExqQZRGRiBGohk3Xls4lJm9MTQeZy5TIXqGJvAIXLguTGONP5PRpChKebgplFsPZKeYxKnCLKaYE4PwehARagM+JuPpHBcTQLjGr1lMc6Sp1k/A5+HEiOViEoFVDUGG7Rbqc0117R0aZ21TDUGfl66W2kxGk6Io5eFGIO4AnhSRn2LNgXgfcHtFV7UIOEHqoO02qgl4mUwkp1xMoSkL4sjpiZx9FXeICG1hqxaiNuBlZX0Qv9dDS73lYjo9x0ymnqEJulrqAOhsqePZvjPztmZFWc7MeoczxvwV8HHgDBABPm6M+Z8VXteCk53FBGSmyo3ZdRC1tkskHPJnWnCoBVE6beEQAyNRqwbCdjk12/MhBsuwIDpbagHoaqnl6JnJzH9PRVHmTtE7nIiE7Z/NQA+WJXEH0GtvmxURuVpEDorIYRH5XJF93iwie0XkeRF5JGv7H4rIfnv7Z9xf0tzIzmICqPF77TTXFHUBbyb9siFrzKi2+y6d1kar3cbAcJTWcBAAn9dDU61/TrUQkYk4kYlEjgWRNtB/Rt1MilIuMz0Cf8/++TSwJ+vlvJ8REfECXwGuwcp6ul5EtuTt0wR8FXiPMeYC4EP29guxWorvBLYB7xKRik61n8piynYxpTLjRh2cWgjQLKa50BYOMjAczbTZcJhrNbUTb8hYECtrc7ZXit2vDPHX975Y0XMoymJT9A5njHmX/XODMWZj1muDMcZNwdxO4LAx5hVjTBy4E7g2b58bgLuMMUfsc520t58P7DbGTBhjklhFeu8r7dJKI9+CyLiY7GlyDk41NaiLaS44o0fPTCQyLiaAlvrgnFxMTsbShpVTFkT29kpx73PHufXfXiGdNhU9j6IsJjMVyl0y04HGmGdm+e51QF/W+37gsrx9NgN+EXkYq+bi74wxtwP7gb8SkRZgEngHRawWEfkE8AmA9evXz7Kk4sSzspjAcjGdHk8wFktmMphALYhycUaPwlThHFgWxEt2y+5S6B2aQAQ6mmsz31Mf9FXcgohMJDAGxuLJTJdfRak2Zspi+psZPjPAlbN8dyEHff7jlg+4FHgrUAM8LiK7jTEvisiXgF8CY0A3kCy4EGNuBW4F2LFjx5wf5/KD1DUBH1HHxZRtQWQLhFoQJZNtNeRaEAF2v1J6DKJncJw14VCmMltE6GyprbgFEbFTc0ejKhBK9TJTodxbyvzufqAj6307cKzAPoPGmHFgXEQexYo5HDLG/DPwzwAi8j/tfStGLN/F5PcyEU8S9HnoqKvN7JcTpFYLomSy4w7Z1kRzXZAzEwmSqTS+EoS3Z2g841Zy6Gqp44XjI+UvdgaGMwKRwHq2UZTqw9X/iSJyoYh8WERucl4uDnsK2CQiG0QkAFyH1Q02m7uBN4qIT0RqsVxQL9rnXG3/XA+8H/i+u0uaGzG7A2jQaz2J1mTFIBpyXEyaxVQO2QKR/ftKuxbizERi2jEz0Ts0kQlMO3S21NJ3eoJkqnKprsMTVrxkZLKgYasoVcGshXIi8kXgzViZSPdiZSX9O7MUyxljkiLyKeB+wAt80xjzvIjcbH9+i+1Kug/YB6SBbxhj9ttf8RM7BpEAPmmMqWj1kxOkDvqnspiiiRRej+QFqafcCY6YKO5xRo9OxFOEs8S2pc5KeR0aj7GqIejqu0aiCYbG4wUtiGTacCwSZX1LbZGjyyOSY0EoSnXippL6g1hun2eNMR8XkVbgG7McA4Ax5l4sUcnedkve+y8DXy5w7BvdnGO+yA9S1/q9JFKGkclErkDkuJjUgpgLq8MhookUIlN/P6dh3+kSMpmO2IHorpbpFgRY7qd8gfjJ0/0MjERZURtgRa0fEaspYzSR5nXntEwTm0Kk0ybLxaQWhFK9uBGISWNMWkSSdvHcSapwLkQ8mcbnkUxBnNPRNW1y3UrZAUlNc50bV2xeNW2oj+NiGiyhFsIphmtfkSsCG1fVA3DoxChvsju8AgxPJPjjH3UX/b53bl3DV26YMXkPsETBGOd3tSCU6sWNQOyxC9q+jlUkNwY8WclFLQbxZDonbdURCIC6rN+dNFcR8M1huI0Cn7vmvGnb5tLy25lvnR3sBqsB4NrGEN39wznb9x2NAPDN39zBljWNnLHjCCG/lz+881kiE+7EKTI5td+IWhBKFeNmotzv27/eYscLwsaYfZVd1sITT+UKRG22QGS5mEJ+T8bSyHaRKOXRVBvAI5RUTT0wEsPvlUwvp2yc6XLZOO8v7Wymscafl0UVcN1N1nEvgRUHUZRqZVYfiYjcLSI3iEidMaanGsUBrFYb2XUNNf7CmUsiQrjGrzUQ84zXIzTXBUqqpj4xEmV1Q6jgmNJtHU0cOT2RIzh7+4bZuKqOxprpdQvhkN+1uyiSlWmlMQilmnFzl/tb4A3ACyLyIxH5oIiEZjvobCOeSmcymKC4BQGWYGgV9fzTXBcoqeX3wHB0mnvJYVt7EwDd/REAjDF090fYbm/PpyHkc32zdzKYRGBkUi0IpXpx0+77EdvNtBGrYvnDWIHqqiKezLMgsgSiPk8gwiG/1kBUgJa6YElDg06MRnOqsbO5qL0RkSm30sBIlFOjMbZ1NBXcvyHkZySawJjZi/GdGog14ZBaEEpV47ZQrgb4ANY86tcCt1VyUYtBLJkm4JsSBWckJkwXiIaQTzOYKkBzfWkdXU8MR1kdLlwzUR/0sWl1fUYgnJ/FBCJc4yORMpnJgjPhuJjam2s1i0mpatwUyv0Aq8L5Pqz23Q8bY6puGovbIDVYozNPjWqR3HyzuiHIg8NRUmmDd5YMsdFogvF4qqgFAZab6VcHTmKMYW/fMH6vcP6ahoL7OtlpI9FEpq9TMSKTCeoCXlrqAhyeQ4PBaudL9x3goQNTToZ3XrSGP3hrRbv1KxXCzWPwt4BzjDE3G2MerEZxAIgnUwSLuZhCuQLxmas28z/ee+GCrW25cMHaRiYTKV4+NftN98RI4RTXbLZ1NHF6PE7/mUm6+yKcvyZM0Ff45u8UQLppnRGZSNBY46ch5NMspjxSacNtj/UQT6VZ31yLMfAPDx6e06wPZfFxE4O4zxiTAhCR2Vp8n7XEknlB6qwsprpArkBsbm3gso0tC7a25cK29kaAaemphTgxYgWzW2ewILbb7qRnjpzhuaPDmcB1IZwCSDcuo+HJBI21ARpCfo1B5HH45BgT8RSfestruPWmHfzDDRcTT6X5wVN9sx+sLDlKdaRXbWS2WJC6xu+d1d2hzA8bV9VTH/SxL6/ArRCZIrkZBOLctgYCPg93PXOUsViyaPwBplKZ3dzwhyfjNNX4M/PJK9kU8GwjP9azubWByzc2853dvaR0uNJZR6kC8f8qsoolQH4ldcBnFcTlu5eUyuH1CBeuC2dSU2diwHYxzWRB+L0eLlwb5pFDpwDY3tFYdN/sGMRsRCYSNNX6SxKV5cLe/ggNIR8bsnpa3bSri6ORyZy4hHJ24KZQrk5EnP1uF5H3iEjVTUjJD1KDZT3kZzAplWVbexMvHh8hlkzNuN+JkSjhkC8nVlTw++wn2fqgj40r64vuV8rNPjKpAlGM7r4I29qbcooX37alldZwkNt39y7iypS54MaCeBQIicg64FfAx4FvV3JRi0G+iwksN5MKxMKytb2JRMpw4PjojPvNVCSXjROH2NreWLDi2sFp4z5bDMIYw/BEgsaaQElWx3IgmkhxYGCUbXmWmt/r4YadnTx66BSvDlZ20p8yv7i5+4kxZkJEfhv4B2PM/xaRZyu9sIUm38UEVqprXVDTWRcS5+bS3R+ZMWZwYjQ2o3sp8312YHrrDAFqsBoyemT2LKbJRIp4Kk1TrZ9wzXQLomdwnH969GX+4toLl12tzPPHhkmlTcFkgOt3dvAPD77EX/z8eXadYyV4bFnTyBs2rVzgVZ69dPdFeOLVocz7re1NXF7hZBlXAiEiu4CPAL9dwnFnFbFkeloK5OqGkKubkDJ/rGuqoaUuQHffMOwqvt+J4SibV89+c+lsqeUvrr2At21pnXE/EaE+6JvVgnCK5JwgNeRaEP/64gm+/2QfH3/9Bja3Fq65qFb29lnJBdsLCPvqcIj3X7KOH+7p56GDVkyoqdbP3v/+9oVc4llLMpXm97/7DEcjk5ltDSEfT/zXt1IbqNzt2M03fwb4PPBTeyLcRuChiq1okShkQXzlI5doS40FRkTY2t7IvhkC1am04dSYOwtCRLhpV5erc4drZk9bdTq5FotBONlVA8PRZScQ+/ojrGkMsbrIf5cvfWArX3z3BQDc/ngvX7rvAJGJOE0FuvEqufzqwEmORib5++sv5q3nraa7L8IN33iCnz17jBsuW1+x87rtxfQeY8yX7GD1oDHm0xVb0SJgjCkYpF7VENR/vIvA1vYmDp8aYyxW+GY9OBYjlTa0uohBlILVj2lmgXAsiHCWBZFtdTjZVc7P5YQToC6GiDW+ty7o4zWrrYSBHnsqoDIz39ndy5rGEO+4sI26oI9d57RwXlsDtz/e46p/2Fxxk8X0PREJi0gd8AJwUEQ+W7EVLQKZedTaoXVJsL2jCWPguSL1EG5qIOaCm8roYXtYUFNNIJMCnR23cCq8TwwvL4GITMTpGZqYMW6UjTMmtndIg9az8cqpMf7tpUFu2Lkenx3XcizjAwOjPN17pmLndnNH3GKMGQHeizVfej3w0YqtaBHIn0etLC5b7YrqYm6mTJuNeRaIsIvK6EwMotaP3+uhxu/NsSCcCu8To8tLIJzpffkZTMXoaK5FBHoG1YKYjTt29+L3CtftzHUlvffitTSEfNz+eOXSh93cEf123cN7gbuNMQmgqkoinQ6eOuNhadBSH2RdU03RimpHIFobC3dynSvhkIsgdVYMAnLnSBhjplxMw+7nWlQD3X0RROCide4EIuT30hYO0XtaLYiZmIgn+fHT/Vxz4RpWNeT+e68N+Pjgpe38y/7jnBqtzL83N3fEfwJ6gDrgURHpBEYqsppFwrEg1MW0dNjW0cjeIj2ZBkaieD1CS938CkRDyDfrAKDIRIKAbTmAHdiOJTKfOf+WTiyzGER3X4TXrKrP1Ia4obOlll6NQczIz549xmg0yU27Ogt+/tHLO0mkDHc+eaQi53cTpP57Y8w6Y8w7jEUv8JaKrGaRiKsFseS4bEMLRyOT7D863YoYGI6xuiE47z2ywjV+xmLJGYN+VqM+f2YeuSUqlgXhuJUagr5lF6R+/tiIa+vBoaulTmMQM2CM4fbHezh/TZhLO1cU3GfjqnreuGkl33vySEV6grkJUjeKyN+KyB779TdY1kTV4ASpVSCWDu+9eB01fi+3P94z7bOTo9GK1Kc0hHykDYzHi7f5cBr1TR0zNcvaCZ5v7WhkcCy2bJr4GWMYHIuxpqm0/yadLXUMjsV16FIR9vSe4cDAKDft6sw8kBTiD9+6if/+ri0z7jNX3NwRvwmMYo0a/TCWe+lb876SRUSD1EuPxho/77tkHXfvPcaZvFkCA8PFR42WQ6Z1xgxuJqdRn0M4KwbhuJW2tVtZWKfGlkccYmQySTJtaC7R5deZyWRSN1Mhbn+8l4aQj2u3r51xvx1dzVxz0ZqKdJ12c0c8xxjzRWPMK/brz7HmU1cNGqRemty0q5NYMs0P9+TOEhgYcdeHqVSm6hqKZzI5w4IcnFnWMBWYdlwtA8sk1XVw3LrulfWl1Qw5AnHktApEPidHo9y3/zgfurSjopXSs+HmjjgpIm9w3ojI64HJGfY/63A6h6pALC3Oawtz2YZm7siaJTARTzIaTRadRV0OU5XRxS2I4UmrUZ9DOOTLFNedGI3SXBego9m68Tkpr9WOMy2uua5UgbA81T0ah5jGnU/2kUgZPlokOL1QuLkj3gx8RUR6RKQH+Efgdyu6qgVmKotJG/MtNT72ui76z0zy8EFrlkCliuRgSiBmKpazWkNkWxA+4sk0sWSKE8NWbMSJjyyXTKYh25VWqkDUB32srA/Qq7UQOSRTab73xBHeuGklG1Yubrh3RttFRLzAjcaYbSISBrCL5qoKTXNdurxtSytt4RD/9Ogr1AV9vHTCagNeCYGYavld2MWUSKUZj6dygtTZxwyMRGkLB2mpC+D3yrLJZBqyLYiV9aVbdZ0tdWpB5PHLF04wMBLlL5fA3PsZBcIYkxKRS+3fq04YHDSLaeni93r46K5Ovnz/Qa67dXdmu+PGmU+mLIjCAjGcVySXc8xkghMj0czcidUNobLabRhjuP7ru7n6gjZ+8/Ub5vw9C8HQmCUQK+bQt6yzpZbHXx6afcciDE8m+OyPuvkvV5/La1YvXnPEnsFx/uznz5OYh8y1l0+Os66phivPWz0PKysPN9GPZ0XkHuBHQEbqjTF3VWxVC4xmMS1tPvGmjVy2oZlYMk0ilaYh5K+IQIRnyWJy2mw0Zt0IG4LWMWcm4gyOxVndYFk2q8PBstptDIxE2f3KaV4+Nc4Nl3Uu6YeX0+NxwiHfnNbY1VLHXc8cJZpIEfKX7uL94VN9PPDCCeqDPv72P2wv+fj54pZHXuaxl4fYWmItSCHaV9Twsdd1VSQrqVTcCEQzMARcmbXNANUnEEv4f8LljN/rYUdXc8XPE/R58HulqItpqlHfdBfTy6esZycnu6otHOLQiZmn4s1Et11Ffmo0xv3PD/DubTOnOi4mg2MxWubgXoLcTKZS26On0oY77DGmv9h3nD995/lzXkc5RCbi/GzvUT5wyTr++v1bF/z8lcRNJfXHC7x+ayEWt1DENAahYHXIDGcVvuWT3ajPwXEx5cdGWsOhsrKYuvuH8XmE9hU13FHBZmzzwenxOC0lBqgdMplMcxhF+sihkxw5PcGn37qJeCrND/LSoReKH+3pJ5pI89HLuxbl/JXETSX1bSLSlPV+hYh8s6KrWmDUglAcGrLSVvPJuJhqCgjEyTGATAZTazjEWCxZdKbFbHT3RThvTQM37erkyZ7THBhYuiHAobF4yRlMDl1l1ELc9lgvqxuCfOotr2HXxha+u/tIJh16oUjbVszOrma2rA0v6LkXAjd3xK3GmIjzxhhzBri4YitaBDRIrTg0zGRBOEHqrDoIp/r6cEYgLBdHm91pdi6prum04bn+Yba1N/HhHR0EfZ4lbUUMjcfn7Nppqg0QDvlKzmTqGRznkUOnuOGy9QR8Hm7a1cnRyCQPHjg5p3XMlUcOneLI6YlFr1eoFG7uiB4RyXSKEpFmqmwmdUyD1IpNuMY3QwwigciU1QBWYz4R6D8zScDryTxJZ2oh5pDJ9MrgOKOxJNvam2iqDfCebWv56bNHZx1mtBik04YzE3N3MQF0rawrud3GHbt78XmEG+wZCU46dKHeXZXktsd7WN0Q5DcuaFvQ8y4Ubm70fwM8JiI/xgpOfxj4q4quaoGJJ9MEvJ6KNLtSzi4agn5Ojoxl3t///AB3PdNPKm04MDBKY40fT1Z2iccj1Ad8jMas6m7n35ATi5hLLYQzKMmZznbTri5+9HQ/dz3dP68pr4NjMb67+wiffMs5mUll+aTThp8808/bL2jLca05DE8mSKUNLSW22cims6WOvX3up6JNxJP8cE8f11y0JjP/2uf1cMNl6/nbXx7iaw+/TMifez0CXLWllfYV5WW/nRqNce9zx0kbQyyZ5uGDp/jMVZuq1vswq0AYY24XkT1YWUwCvN8Y84KbLxeRq4G/A7zAN4wx/6vAPm8G/i/gx5p3fYW9/Y+A/4glSs8BHzfGVKTyKJ6cPo9aWZ5kDwAC+OpDh3llcJz1zbW01Ad519bp2UQNIUsgsov3WssQiO6+CLUBb2Zu80XtjZyzqo5/Pzw0rwLx1Yde5pu/fpXLNjZz+caWgvs8fOgkn/3xPl46OcZ/fcf50z4fGp9bFXU265truPe546TSxlVq52OHhxiNJrnutR0526/b2cHXH32FL913oOBxjxw6xbc+vnPO6wT46395kbueOZp5Xx/0ZayYasSVq8gWBFei4GBXYX8FeBvQDzwlIvdki4sd/P4qcLUx5oiIrLa3rwM+jTXudFJEfghcB3y7lDW4JZZMaQaTAtgDgGxXTiyZ4oXjI/z2GzbyuWvOm/GYY8O5Lcjrgj4agj5OziGTqbt/mAvXNebcLM9ZVT+vFceT8RQ/ftrK+unuixQViG8/ZsU+fvBUH3901WZqArm1Ck6R3FyqqB3awiFSaatluJs27t39Ebwe4eL1TTnbVzeEeOq/XUU0Mb1d+62PvsLXHnmZI0MTrG+ZmxVxejzOL/Yd5/qdHfzJ1da/h5DfO6f6jbOFSt4VdwKH7Q6wceBO4Nq8fW4A7jLGHAEwxmRHmHxAjYj4gFrgWKUWqhaE4tAQ8jEeT5FMpXnx+CiJlGFb+8zFT05MIv/m1toYKrmjazyZ5oVjI9PO6fjp0/OUpfPzfccYiSYJeD1FR7u+cmqMRw+d4orNqxieTPCzvUen7TM0x0Z92WSsLZd/q+7+YTa3NhTschrye2mqDUx73bSrC48I33li7sH+HzzVRzyZ5uOv35D53moWB6isQKwDshOT++1t2WwGVojIwyLytIjcBGCMOQr8H+AIcBwYNsY8UOgkIvIJZ5jRqVOn5rTQeEoFQrFwspLGYslpsYDZjmnLm5HdGg6W7GI6ODBKPJWeds7OllpiyXRZ1dnZfHd3L5tW13PVltV029eZz+2P9+L3Cl/+0FbOXxPmtsd6pk3bcwSinBiEU1zo5m9ljKG7LzKraBc6x29c0MoP9/QVtDBmI5U2fGd3L5dtaC65oO9sppJ3xULOxPzHHx9wKfBO4DeAL4jIZjtr6lpgA7AWqBORGwudxBhzqzFmhzFmx6pVq+a0UCdIrSjhTMvvJHv7IqxqCLJmltkTRS2IcKjkNNe9jii1N+Vs72y2CsrmY7jOvv4I3f3D3Hh5J9vam+g/M5npyOowFkvyk6f7ecdFa1jdEOJjuzo5MDDKk6+eztnPOW4ufZgc2krofts7NMHwZGJW0S7EjZd3EplI8PPu0p0RDx88ydHIJDft6ir52LOZSt4V+4HsKFI7091E/cB9xphxY8wg8CiwDbgKeNUYc8oYk8Bq6/G6Si1UXUyKQ2aqXDSReVKdLbvN6eGULxBt4RAnR2MluYX29UVorgvQvqImZ/vU9LXy4xDf2d1Ljd/L+y5Zl7nR5ruZfvpMP6OxJB97XRcA125fR2ONn9vy0kiHxqz25/4yHrBa6q354m5cTN1FBNQNuza2sGl1faY9Rync/ngvreEgb7+gteRjz2YqeVd8CtgkIhtEJIAVZL4nb5+7gTeKiE9EaoHLgBexXEuXi0itWP93vtXeXhFiKhCKjWNBHItEeWVw3NWNyLEg8luQtzrB13H3gep9/cNsLSBKa5tq8HuFnjItiOGJBPd0H+O9F68jHPJz4bpGRGCv3fsJLDfObY/3srW9kYttAakJeLnutR3c//wJjkWm5oWdHp97FbWD1yOsbnDnjuvuGybk97C5tb7k84gIH93Vyb7+YR57eZDIRNzV6/ljwzxy6BTX71xflhCejVSs4M0YkxSRTwH3Y6W5ftMY87yI3Gx/fosx5kURuQ/YB6SxUmH3A9h1F88ASeBZ4NZKrTWeTGsWkwJMNd/79eFBjJk9/gCwvrmWhpBv2hhU5/3xSDTT5XUmxmNJXjo5ym9cOL3oyusROppry7Yg7n9+gGgizUcus1Iz64M+Nq2uz8RbAB57eYjDJ8f4Px/aliNUN17eyT89+gr3dB/j5ivOAaxaipUlzqIuhFt3XHd/hAvXNhat25iN9128ji/9ywFu+PoTJR2XXZS3nKhoRbQx5l7g3rxtt+S9/zLw5QLHfhH4YiXX5xBLpWkMTC8CUpYfjjXw74cHAdjqIhj6oR0dXH1h27SMloxb6PSEK6F57ugwaUPmqT2frpY6esqcvnY0MokInNc2FWjd2t7EQwdOYoxBRLj98R6a6wK8a+uanGM7mmvpaqll75FIZtvp8TjnrCr9aT6ftnCIl07O3P02kUrz/LFhPnLZ3NtaNIT8fOvjO3n+WOHMrWKcs6o+U5S3nKiqlhlzRYPUikN2b6WullqaXARfvR4puN96e2ZFr8tOpc5TfDFRWt9cyxOvDGVu5HNhaDxGU40/5wl8W0cTP366n/4zk3g9wi9fOMEn3nROwRTObR1NOYHqofE4OzeU52ICy9pyRLkYh06MEk2kXYn2TOzc0MzODZVvH18N6F0RiGuhnGKT3Wdp6xwCodnUBny0hoP0uuxU2t03TEdzTdHGd10ttYzHUwzaxWlz4XSBxnrb7evc1z/M9544goGMCyqfre1NHB+OcnIkSmoe+jA5uOl+291nPfVvn0MGkzI39K6I1kEoU/i9HmrsJ+e5pFLm09lS5zpusLcvMmNQvNMeYH/k9NzjEIMFWnOf29ZAwOvhqZ7T3PnUEa48d3XRiX3bO6yn9+7+Yc5MxDGGeRnS46b77b7+CE21/oxlplQevSsCsYS6mJQpHCvCuRmWQ2dzravMo1OjMY5GJmcUiK7McJ25xyGGxmKszCtqC/g8bFkb5vtPHmFwLD5j6+oL1lotQLr7IpyehypqBzfdb/f2Rdja3qRNNRcQvStiWRBBv/4pFIuGkA+vR9iypnyB6FpZx6nRGOOzDA5yU7W9rqkGr0fKymSypr9Nf+Lf1t5ILJmmq6WWN20qXnAa8ns5t7WB7v4Ig3aRXDlV1A6zdb+diCc5dGKU7WXGH5TS0CA1GqRWcllRGyDo805rTDcXpgrcJjITx4bGYlx3627+1we2cmmnNWqluy+CR+DCdcWnkgV8HtY2heZcC5FMpTkzkSj4xL+towke7+XGyztz2pkXYltHE/9v37FMo75CglMqhdpt7D86zJ/8ZB+JVJp4Mk3aZdqxMn/oXRH45FtewxXnzq1Nh1J9fOFdW/jSB+Zn+LzjFsqOGzzx6mleOjnG1x5+ObNt7wwN6PK/b64WxBl7ZGq+iwng7Re08ekrX8P1LnL9t3c0MhJN8swRa4bDfFgQtQEfDSFfjovpX/Yf58DAKOesquf8NWE+vKOdXecU7jqrVAa1ILAEQlEc5vMp1Wktnf3U321XLf/qwAn6Tk/QvqKGff0RrnYxlayzpZafdx+f01qmZjdMf+KvD/r4T28/19X3ONldDx04iUh5fZiyaQuHciyI7r5hzmtr4Gs3Xjov36+UjloQilJBwiE/LXWBnKf+vX0ROpprMu2nj5yeIDLhrgFdV0sdw5MJIhOlp7pmXEJlPvFvWl1Pjd9Lz9AEK2oDrob8uKGtMcSAPT8jnTZ090fUpbTIqEAoSoXpbKnNZB6l0obnjg5z5bmrefuWVn7wVB9PvGIVnrnp+9TZMveurk5r7kIuplLweT1ctM4KFs9HBpNDaziUcTG9OjTOaDSZqdFQFgcVCEWpMNm1EIdPjjERT7Gto4mbdnURmUjw//3rIdcN6DozLqvS4xBOa+5CLqZScaqZ56NIzqEtHOLUWIxU2mTccGpBLC4qEIpSYTpbajk+EiWaSOXc+C7f2Mzm1nqOD0e5aJ27BnSZ9h1zsSDG4ngEmmrK7zvm3LjnI0Dt0No4NXo0fy63sjioQChKhelqqcMY6D8zwd7+CA0hHxta6uz2012A+7YeIb+XNY0hfrb3KH/60+f4H794gYcPnpy239O9Z7jtsZ6cbUN2a+7Z0ljd4LS7mI8UV4e2rNGj3f3DXJQ3l1tZeFQgFKXCZNxCgxP2EKKmzE36/Rev402bV/HOvM6pM/Hei9cRS6S5b/8At+/u5Q/v3MtEfKoQzxjDf797P3/5ixdIptKZ7UNjsXm7obevqOGt563m9a+Zv7RTRyD6z0xac7nVvbToaJqrolQYpxbi4IlRDgyMcvMVGzOf1QV93P5bO0v6vj+5+jz+5OrzAHiq5zQfuuVxfvrs0Uwb7D29Z3j+2AhgDT5yUm2tRn3z4xISEf75N187L9/l0Gr3Y3r44ElrLrcGqBcdtSAUpcI01foJh3zc+9xxUmlTdpfYbHZ0ruDCdWG+9esejLFGm3771z2Zz7OD2UPzMP2tkrTUWaNH//XFEwBsm4deWEp5qEAoSoUREbpW1mWe6uezXbWI8Fuv38Dhk2P820uDHB+e5L7nB7h2+1ogd4b14FiMlfPQebVSOKNHz0wkWFkfYF1TzewHKRVFBUJRFgAn+6gtHMp0Lp0v3rl1DSvrg3zz16/ynd29pI3hj992LiG/J1PBHU+mGY0m5zUttRI4f5tt2rV1SaACoSgLgBOHqITbJOjzcuPl63n44Cluf6yXq85vZX1LbU7fpkxr7nlMS60ETqBaA9RLAxUIRVkAnEymSt34PnJZJwGvh9FYkt98XVfmnI4FkWnNPY9pqZXA6eqqArE0UIFQlAXgovZGPAKvO2dlRb5/VUOQGy5bz8Xrm3id3fG0q6WOI0MTpNImY0HMZ2FbJdiwso6Az8M2nfuwJNA0V0VZAM5rC7P3i28nHCq/irkYf/aeC3Ler2+pJZ5KMzASzXRyXeoxiOt2dvDmc1fRNE8dYpXyUAtCURaISopDIZy4R+/geFYn16XtYgr6vJmGhMriowKhKFVKZprd6QmGxuP4vUI4pE4DxT36r0VRqpQ1jTUEvB56hsY5YxfJaeqoUgoqEIpSpXg9QkdzDb2DEyTT6Xlp860sL1QgFKWK6Wqpo2donJDfW/agIGX5oTEIRalirGFFEwyOxZZ8BpOy9FCBUJQqpmtlLZOJFEcjk+piUkpGBUJRqhgnZdSYpV8kpyw9VCAUpYrpslNdYekXySlLDxUIRali1jXV4LOn1y31Ijll6aECoShVjM/roX2FNVdBXUxKqahAKEqV48Qh1MWklIoKhKJUOU7LDXUxKaWihXKKUuW89+J1eD1CXcC72EtRzjIqakGIyNUiclBEDovI54rs82YR2Ssiz4vII/a2c+1tzmtERD5TybUqSrVyyfoVfPHdF2gfJqVkKmZBiIgX+ArwNqAfeEpE7jHGvJC1TxPwVeBqY8wREVkNYIw5CGzP+p6jwE8rtVZFURRlOpW0IHYCh40xrxhj4sCdwLV5+9wA3GWMOQJgjDlZ4HveCrxsjOmt4FoVRVGUPCopEOuAvqz3/fa2bDYDK0TkYRF5WkRuKvA91wHfr9AaFUVRlCJUMkhdyOFpCpz/UiwroQZ4XER2G2MOAYhIAHgP8PmiJxH5BPAJgPXr18/DshVFURSorAXRD3RkvW8HjhXY5z5jzLgxZhB4FNiW9fk1wDPGmBPFTmKMudUYs8MYs2PVqlXztHRFURSlkgLxFLBJRDbYlsB1wD15+9wNvFFEfCJSC1wGvJj1+fWoe0lRFGVRqJiLyRiTFJFPAfcDXuCbxpjnReRm+/NbjDEvish9wD4gDXzDGLMfwBaMtwG/W6k1KoqiKMURY/LDAmcvO3bsMHv27FnsZSiKopw1iMjTxpgdBT+rJoEQkVNAKemwK4HBCi1nqbIcrxmW53Uvx2uG5Xnd5VxzpzGmYAC3qgSiVERkTzHlrFaW4zXD8rzu5XjNsDyvu1LXrM36FEVRlIKoQCiKoigFWe4CcetiL2ARWI7XDMvzupfjNcPyvO6KXPOyjkEoiqIoxVnuFoSiKIpSBBUIRVEUpSDLUiDcDDKqBkSkQ0QeEpEX7YFMf2hvbxaRX4rIS/bPFYu91vlGRLwi8qyI/MJ+vxyuuUlEfiwiB+z/5ruq/bpF5I/sf9v7ReT7IhKqxmsWkW+KyEkR2Z+1reh1isjn7fvbQRH5jbmed9kJRNYgo2uALcD1IrJlcVdVMZLAHxtjzgcuBz5pX+vngF8ZYzYBv7LfVxt/SG5fr+VwzX+H1fzyPKymly9SxdctIuuATwM7jDEXYrX0uY7qvOZvA1fnbSt4nfb/49cBF9jHfNW+75XMshMI3A0yqgqMMceNMc/Yv49i3TDWYV3vbfZutwHvXZQFVggRaQfeCXwja3O1X3MYeBPwzwDGmLgxJkKVXzdWP7kaEfEBtVgdo6vumo0xjwKn8zYXu85rgTuNMTFjzKvAYaz7XsksR4FwM8io6hCRLuBi4Amg1RhzHCwRAVYv4tIqwf8F/gtWA0iHar/mjcAp4Fu2a+0bIlJHFV+3MeYo8H+AI8BxYNgY8wBVfM15FLvOebvHLUeBcDPIqKoQkXrgJ8BnjDEji72eSiIi7wJOGmOeXuy1LDA+4BLga8aYi4FxqsO1UhTb534tsAFYC9SJyI2Lu6olwbzd45ajQLgZZFQ1iIgfSxy+a4y5y958QkTW2J+vAQrNAj9beT3wHhHpwXIfXiki36G6rxmsf9f9xpgn7Pc/xhKMar7uq4BXjTGnjDEJ4C7gdVT3NWdT7Drn7R63HAXCzSCjqkBEBMsn/aIx5m+zProH+Jj9+8ewBjdVBcaYzxtj2o0xXVj/bR80xtxIFV8zgDFmAOgTkXPtTW8FXqC6r/sIcLmI1Nr/1t+KFWer5mvOpth13gNcJyJBEdkAbAKenNMZjDHL7gW8AzgEvAz86WKvp4LX+QYs03IfsNd+vQNowcp6eMn+2bzYa63Q9b8Z+IX9e9VfM7Ad2GP/9/4ZsKLarxv4c+AAsB+4AwhW4zVjTdY8DiSwLITfnuk6gT+1728HgWvmel5ttaEoiqIUZDm6mBRFURQXqEAoiqIoBVGBUBRFUQqiAqEoiqIURAVCURRFKYgKhKIoilIQFQhFmSdEZK2I/NjFfmNFtn9bRD44/ytTlLmhAqEo84Qx5pgxZlFu8HY3U0WZV1QglGWFiHTZw3S+bg+aeUBEaors+7CIfElEnhSRQyLyRnu7V0S+LCJPicg+EfndrO/eb/9eKyI/tD//gYg8ISI7sr77r0SkW0R2i0hr1mmvEpF/s8/3LnvfkIh8S0Seszu1vsXe/psi8iMR+TnwgIisEZFHRWSvPUDnjZX5KyrLBRUIZTmyCfiKMeYCIAJ8YIZ9fcaYncBngC/a234bq7X0a4HXAr9j97zJ5veBM8aYrcBfApdmfVYH7DbGbAMeBX4n67Mu4AqseRa3iEgI+CSAMeYi4HrgNns7wC7gY8aYK4EbgPuNMduxBgbtne0PoSgzoWapshx51Riz1/79aaybcjHuKrDf24GtWfGCRizROZR13BuwJrxhjNkvIvuyPosDv8j63rdlffZDY0waeElEXgHOs7/rH+zvOiAivcBme/9fGmOcQTJPAd+0O/j+LOsaFWVOqAWhLEdiWb+nmPlBKVZgPwH+wBiz3X5tMNagmmwK9eR3SJipJmj5589vjmZm+a7xzI7W1LE3AUeBO0TkphmOU5RZUYFQlNK5H/g9+0kdEdlsT2/L5t+BD9ufbwEucvndHxIRj4icgzUl7iCWG+ojzrmA9fb2HESkE2tY0tex2rxfUuqFKUo26mJSlNL5Bpa76Rl7DsEpps89/ipWrGAf8CxWC+5hF999EHgEaAVuNsZEReSrWPGI54Ak8JvGmJh16hzeDHxWRBLAGKAWhFIW2u5bUSqAiHgBv32DPwerX/9mY0x8kZemKK5RC0JRKkMt8JDthhLg91QclLMNtSCUZY+IfAVrlnU2f2eM+dZirEdRlgoqEIqiKEpBNItJURRFKYgKhKIoilIQFQhFURSlICoQiqIoSkH+f3UvttdtOkYvAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "crossval_score = {}\n",
    "scores = []\n",
    "for n in tqdm(range(1, 101)):\n",
    "    model = KNeighborsClassifier(n_neighbors = n)\n",
    "    scores.append(cross_validate(model, X_train, y_train)['test_score'].mean())\n",
    "crossval_score['KNN'] = (max(scores), np.argmax(np.asarray(scores))+1)\n",
    "plt.plot(range(1, 101), scores)\n",
    "plt.xlabel('n_neighbors')\n",
    "plt.ylabel('cross-validation score')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "s8Jl8rv416DK"
   },
   "source": [
    "## Дополнительное моделирование\n",
    "### Задание 7 (2 балла)\n",
    "Теперь более серьёзно подойдём к моделированию. Попробуйте несколько алгоритмов из тех, что мы проходили в курсе. Вам помогут ноутбуки с линейными алгоритмами и выбором модели. Хотя бы для одного алгоритма проведите подбор оптимального гиперпараметра. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "id": "iHSIbb0g16DK"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 10/10 [00:02<00:00,  4.47it/s]\n"
     ]
    },
    {
     "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": [
    "from sklearn.svm import SVC\n",
    "\n",
    "scores=[]\n",
    "for n in tqdm(range(1, 11)):\n",
    "    model = SVC(kernel='poly', degree=n)\n",
    "    scores.append(cross_validate(model, X_train, y_train)['test_score'].mean())\n",
    "crossval_score['SVM'] = (max(scores), np.argmax(np.asarray(scores))+1)\n",
    "plt.plot(range(1, 11), scores)\n",
    "plt.xlabel('degree')\n",
    "plt.ylabel('cross-validation score')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [00:24<00:00,  4.15it/s]\n"
     ]
    },
    {
     "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": [
    "from sklearn.ensemble import RandomForestClassifier, GradientBoostingClassifier\n",
    "\n",
    "scores = []\n",
    "for n in tqdm(range(1, 101)):\n",
    "    model = RandomForestClassifier(n_estimators = n)\n",
    "    scores.append(cross_validate(model, X_train, y_train)['test_score'].mean())\n",
    "crossval_score['RandomForest'] = (max(scores), np.argmax(np.asarray(scores))+1)\n",
    "plt.plot(range(1, 101), scores)\n",
    "plt.xlabel('n_estimators')\n",
    "plt.ylabel('cross-validation score')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "100%|██████████| 100/100 [00:15<00:00,  6.58it/s]\n"
     ]
    },
    {
     "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": [
    "scores = []\n",
    "for n in tqdm(range(1, 101)):\n",
    "    model = GradientBoostingClassifier(n_estimators = n)\n",
    "    scores.append(cross_validate(model, X_train, y_train)['test_score'].mean())\n",
    "crossval_score['GradientBoosting'] = (max(scores), np.argmax(np.asarray(scores))+1)\n",
    "plt.plot(range(1, 101), scores)\n",
    "plt.xlabel('n_estimators')\n",
    "plt.ylabel('cross-validation score')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "VPcw62S216DK"
   },
   "source": [
    "## Результаты моделирования\n",
    "\n",
    "### Задание 8 (1 балл)\n",
    "Измерьте качество итоговой модели на кросс-валидации. Выполните предсказание на тестовом множестве и сохраните их в переменную ``y_test``. Измерьте итоговое качество на тестовом множестве."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "id": "N5EYIrRE16DL"
   },
   "outputs": [
    {
     "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": [
    "models = crossval_score.keys()\n",
    "plt.bar(models, [crossval_score[m][0] for m in models])\n",
    "plt.xlabel('Models')\n",
    "plt.ylabel('Cross-validation score')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "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": [
    "from sklearn.metrics import roc_auc_score\n",
    "\n",
    "roc_auc = []\n",
    "model = KNeighborsClassifier(n_neighbors = crossval_score['KNN'][1])\n",
    "model.fit(X_train, y_train)\n",
    "roc_auc.append(roc_auc_score(y_test, model.predict_proba(X_test)[:,1]))\n",
    "model = SVC(kernel='poly', degree=crossval_score['SVM'][1], probability=True)\n",
    "model.fit(X_train, y_train)\n",
    "roc_auc.append(roc_auc_score(y_test, model.predict_proba(X_test)[:,1]))\n",
    "model = RandomForestClassifier(n_estimators = crossval_score['RandomForest'][1])\n",
    "model.fit(X_train, y_train)\n",
    "roc_auc.append(roc_auc_score(y_test, model.predict_proba(X_test)[:,1]))\n",
    "model = GradientBoostingClassifier(n_estimators = crossval_score['GradientBoosting'][1])\n",
    "model.fit(X_train, y_train)\n",
    "roc_auc.append(roc_auc_score(y_test, model.predict_proba(X_test)[:,1]))\n",
    "plt.bar(models, roc_auc)\n",
    "plt.xlabel('Models')\n",
    "plt.ylabel('ROC-AUC score')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "e6n93nVF16DL"
   },
   "source": [
    "## Выводы\n",
    "### Задание 9 (3 балла)\n",
    "Лучший результат показали градиентный бустинг и случайный лес.\n",
    "Метод ближайших соседей работает плохо, поскольку параметры пассажиров не являются однородными.\n",
    "Метод опроных векторов работает плохо, поскольку множество выживших плохо приближается алгебраическим."
   ]
  }
 ],
 "metadata": {
  "colab": {
   "name": "homework_module_1.ipynb",
   "provenance": []
  },
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.9.12"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}