{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Основы анализа данных в Python\n",
    "\n",
    "*Алла Тамбовцева*\n",
    "\n",
    "## Практикум по линейной регрессии: часть 1\n",
    "\n",
    "### Подготовка к работе"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Импортируем необходимые библиотеки:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Зафиксируем настройки графиков `seaborn` – стиль и размер изображений:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "sns.set(style = \"whitegrid\", rc = {'figure.figsize': (16, 9)}) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Загрузим данные из знакомого файла `c.xlsx` по видеоиграм:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>Name</th>\n",
       "      <th>Platform</th>\n",
       "      <th>Year_of_Release</th>\n",
       "      <th>Genre</th>\n",
       "      <th>Publisher</th>\n",
       "      <th>NA_Sales</th>\n",
       "      <th>EU_Sales</th>\n",
       "      <th>JP_Sales</th>\n",
       "      <th>Other_Sales</th>\n",
       "      <th>Global_Sales</th>\n",
       "      <th>Critic_Score</th>\n",
       "      <th>Critic_Count</th>\n",
       "      <th>User_Score</th>\n",
       "      <th>User_Count</th>\n",
       "      <th>Developer</th>\n",
       "      <th>Rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>Wii Sports</td>\n",
       "      <td>Wii</td>\n",
       "      <td>2006.0</td>\n",
       "      <td>Sports</td>\n",
       "      <td>Nintendo</td>\n",
       "      <td>41.36</td>\n",
       "      <td>28.96</td>\n",
       "      <td>3.77</td>\n",
       "      <td>8.45</td>\n",
       "      <td>82.53</td>\n",
       "      <td>76.0</td>\n",
       "      <td>51.0</td>\n",
       "      <td>8</td>\n",
       "      <td>322.0</td>\n",
       "      <td>Nintendo</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>Super Mario Bros.</td>\n",
       "      <td>NES</td>\n",
       "      <td>1985.0</td>\n",
       "      <td>Platform</td>\n",
       "      <td>Nintendo</td>\n",
       "      <td>29.08</td>\n",
       "      <td>3.58</td>\n",
       "      <td>6.81</td>\n",
       "      <td>0.77</td>\n",
       "      <td>40.24</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>Mario Kart Wii</td>\n",
       "      <td>Wii</td>\n",
       "      <td>2008.0</td>\n",
       "      <td>Racing</td>\n",
       "      <td>Nintendo</td>\n",
       "      <td>15.68</td>\n",
       "      <td>12.76</td>\n",
       "      <td>3.79</td>\n",
       "      <td>3.29</td>\n",
       "      <td>35.52</td>\n",
       "      <td>82.0</td>\n",
       "      <td>73.0</td>\n",
       "      <td>8.3</td>\n",
       "      <td>709.0</td>\n",
       "      <td>Nintendo</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>Wii Sports Resort</td>\n",
       "      <td>Wii</td>\n",
       "      <td>2009.0</td>\n",
       "      <td>Sports</td>\n",
       "      <td>Nintendo</td>\n",
       "      <td>15.61</td>\n",
       "      <td>10.93</td>\n",
       "      <td>3.28</td>\n",
       "      <td>2.95</td>\n",
       "      <td>32.77</td>\n",
       "      <td>80.0</td>\n",
       "      <td>73.0</td>\n",
       "      <td>8</td>\n",
       "      <td>192.0</td>\n",
       "      <td>Nintendo</td>\n",
       "      <td>E</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>Pokemon Red/Pokemon Blue</td>\n",
       "      <td>GB</td>\n",
       "      <td>1996.0</td>\n",
       "      <td>Role-Playing</td>\n",
       "      <td>Nintendo</td>\n",
       "      <td>11.27</td>\n",
       "      <td>8.89</td>\n",
       "      <td>10.22</td>\n",
       "      <td>1.00</td>\n",
       "      <td>31.37</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   Unnamed: 0                      Name Platform  Year_of_Release  \\\n",
       "0           0                Wii Sports      Wii           2006.0   \n",
       "1           1         Super Mario Bros.      NES           1985.0   \n",
       "2           2            Mario Kart Wii      Wii           2008.0   \n",
       "3           3         Wii Sports Resort      Wii           2009.0   \n",
       "4           4  Pokemon Red/Pokemon Blue       GB           1996.0   \n",
       "\n",
       "          Genre Publisher  NA_Sales  EU_Sales  JP_Sales  Other_Sales  \\\n",
       "0        Sports  Nintendo     41.36     28.96      3.77         8.45   \n",
       "1      Platform  Nintendo     29.08      3.58      6.81         0.77   \n",
       "2        Racing  Nintendo     15.68     12.76      3.79         3.29   \n",
       "3        Sports  Nintendo     15.61     10.93      3.28         2.95   \n",
       "4  Role-Playing  Nintendo     11.27      8.89     10.22         1.00   \n",
       "\n",
       "   Global_Sales  Critic_Score  Critic_Count User_Score  User_Count Developer  \\\n",
       "0         82.53          76.0          51.0          8       322.0  Nintendo   \n",
       "1         40.24           NaN           NaN        NaN         NaN       NaN   \n",
       "2         35.52          82.0          73.0        8.3       709.0  Nintendo   \n",
       "3         32.77          80.0          73.0          8       192.0  Nintendo   \n",
       "4         31.37           NaN           NaN        NaN         NaN       NaN   \n",
       "\n",
       "  Rating  \n",
       "0      E  \n",
       "1    NaN  \n",
       "2      E  \n",
       "3      E  \n",
       "4    NaN  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "games = pd.read_excel(\"c.xlsx\")\n",
    "games.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Удалим ненужные столбцы и сохраним изменения:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "games.drop(columns = [\"Unnamed: 0\", \"Critic_Count\", \"User_Count\"], inplace = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Часть 1: визуализация связей"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Построим корреляционную матрицу:"
   ]
  },
  {
   "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>Year_of_Release</th>\n",
       "      <th>NA_Sales</th>\n",
       "      <th>EU_Sales</th>\n",
       "      <th>JP_Sales</th>\n",
       "      <th>Other_Sales</th>\n",
       "      <th>Global_Sales</th>\n",
       "      <th>Critic_Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Year_of_Release</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.092562</td>\n",
       "      <td>0.003842</td>\n",
       "      <td>-0.168386</td>\n",
       "      <td>0.037700</td>\n",
       "      <td>-0.076433</td>\n",
       "      <td>0.011411</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NA_Sales</th>\n",
       "      <td>-0.092562</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.765336</td>\n",
       "      <td>0.449598</td>\n",
       "      <td>0.638654</td>\n",
       "      <td>0.941010</td>\n",
       "      <td>0.240755</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>EU_Sales</th>\n",
       "      <td>0.003842</td>\n",
       "      <td>0.765336</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.435068</td>\n",
       "      <td>0.722796</td>\n",
       "      <td>0.901239</td>\n",
       "      <td>0.220752</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>JP_Sales</th>\n",
       "      <td>-0.168386</td>\n",
       "      <td>0.449598</td>\n",
       "      <td>0.435068</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.291096</td>\n",
       "      <td>0.612300</td>\n",
       "      <td>0.152593</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Other_Sales</th>\n",
       "      <td>0.037700</td>\n",
       "      <td>0.638654</td>\n",
       "      <td>0.722796</td>\n",
       "      <td>0.291096</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.749242</td>\n",
       "      <td>0.198554</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Global_Sales</th>\n",
       "      <td>-0.076433</td>\n",
       "      <td>0.941010</td>\n",
       "      <td>0.901239</td>\n",
       "      <td>0.612300</td>\n",
       "      <td>0.749242</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.245471</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Critic_Score</th>\n",
       "      <td>0.011411</td>\n",
       "      <td>0.240755</td>\n",
       "      <td>0.220752</td>\n",
       "      <td>0.152593</td>\n",
       "      <td>0.198554</td>\n",
       "      <td>0.245471</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 Year_of_Release  NA_Sales  EU_Sales  JP_Sales  Other_Sales  \\\n",
       "Year_of_Release         1.000000 -0.092562  0.003842 -0.168386     0.037700   \n",
       "NA_Sales               -0.092562  1.000000  0.765336  0.449598     0.638654   \n",
       "EU_Sales                0.003842  0.765336  1.000000  0.435068     0.722796   \n",
       "JP_Sales               -0.168386  0.449598  0.435068  1.000000     0.291096   \n",
       "Other_Sales             0.037700  0.638654  0.722796  0.291096     1.000000   \n",
       "Global_Sales           -0.076433  0.941010  0.901239  0.612300     0.749242   \n",
       "Critic_Score            0.011411  0.240755  0.220752  0.152593     0.198554   \n",
       "\n",
       "                 Global_Sales  Critic_Score  \n",
       "Year_of_Release     -0.076433      0.011411  \n",
       "NA_Sales             0.941010      0.240755  \n",
       "EU_Sales             0.901239      0.220752  \n",
       "JP_Sales             0.612300      0.152593  \n",
       "Other_Sales          0.749242      0.198554  \n",
       "Global_Sales         1.000000      0.245471  \n",
       "Critic_Score         0.245471      1.000000  "
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# numeric_only = True для новых версий\n",
    "\n",
    "games.corr()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Визуализируем корреляционную матрицу – построим тепловую карту (*heatmap*):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x648 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# cmap: color map, цветовая палитра, варианты: \n",
    "# https://matplotlib.org/stable/gallery/color/colormap_reference.html\n",
    "# annot = True: включаем подписи самих коэффициентов Пирсона\n",
    "\n",
    "sns.heatmap(games.corr(), cmap=\"YlGnBu\", annot = True);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Сфокусируемся на конкретных переменных, `Global_Sales` и `NA_Sales`, и построим диаграмму рассеивания с наложенной на нее регрессионной прямой (парная линейная регрессия):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lmplot(data = games, x=\"NA_Sales\", y=\"Global_Sales\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Избавимся от не совсем типичного наблюдения:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "games_upd = games[games[\"NA_Sales\"] < 40]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Перестроим график:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.lmplot(data = games_upd, x=\"NA_Sales\", y=\"Global_Sales\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Попробуем другую модель:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# order=3: полиномиальная модель степени 3\n",
    "# в нее включена константа, NA_Sales, \n",
    "# NA_Sales в степени 2 и NA_Sales в степени 3\n",
    "\n",
    "sns.lmplot(data = games_upd, x=\"NA_Sales\", y=\"Global_Sales\", order = 3);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Часть 2: парная и множественная линейная регрессия, вся выборка является обучающей"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Импортируем необходимые функции:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.metrics import mean_squared_error, mean_absolute_error, r2_score"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Задача 1 (из демоверсии независимого экзамена)\n",
    "\n",
    "Оцените линейную регрессию, уравнение которой имеет следующий вид:\n",
    "\n",
    "$$\n",
    "\\widehat{\\text{Global_Sales}}_i=\\hat{\\omega}_0+\\hat{\\omega}_1 * \\text{NA_Sales}_i,\n",
    "$$\n",
    "\n",
    "где нижние индексы обозначают значения соответствующих переменных для наблюдения $i$.\n",
    "\n",
    "В ответ выпишите оценку коэффициента при свободном члене.\n",
    "\n",
    "Пример ответа:  0.06"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# создаем объект класса LinearRegression\n",
    "# по умолчанию реализует OLS модель\n",
    "\n",
    "lm = LinearRegression()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.06204106257017372\n",
      "[1.79053279]\n"
     ]
    }
   ],
   "source": [
    "Y = games[\"Global_Sales\"]\n",
    "X = games[[\"NA_Sales\"]]\n",
    "\n",
    "model01 = lm.fit(X, Y)\n",
    "print(model01.intercept_) # ответ 0.06\n",
    "print(model01.coef_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Задача 2 (из демоверсии независимого экзамена)\n",
    "\n",
    "Оцените линейную регрессию, уравнение которой имеет следующий вид:\n",
    "\n",
    "$$\n",
    "\\widehat{\\text{Global_Sales}}_i=\\hat{\\omega}_0+\\hat{\\omega}_1 * \\text{NA_Sales}_i,\n",
    "$$\n",
    "\n",
    "где нижние индексы обозначают значения соответствующих переменных для наблюдения $i$.\n",
    "\n",
    "В ответ выпишите значение среднеквадратичной ошибки (MSE) на обучающей выборке.\n",
    "\n",
    "Пример ответа: 12.00"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE: 0.2743356033079392\n",
      "MAE: 0.19745859404243862\n",
      "R2: 0.8855007503060843\n"
     ]
    }
   ],
   "source": [
    "Y_pred = model01.predict(X)\n",
    "\n",
    "print(\"MSE:\", mean_squared_error(Y, Y_pred)) # ответ 0.27\n",
    "print(\"MAE:\", mean_absolute_error(Y, Y_pred))\n",
    "print(\"R2:\", r2_score(Y, Y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Задача 3 (из тренировочной версии независимого экзамена)\n",
    "\n",
    "Оцените линейную регрессию, уравнение которой имеет следующий вид:\n",
    "\n",
    "$$\n",
    "\\widehat{\\text{Other_Sales}}_i=\\hat{\\omega}_0+\\hat{\\omega}_1 * \\text{NA_Sales}_i + \\hat{\\omega}_2 * \\text{JP_Sales}_i,\n",
    "$$\n",
    "\n",
    "где нижние индексы обозначают значения соответствующих переменных для наблюдения $i$.\n",
    "\n",
    "В ответ выпишите оценку коэффициента при свободном члене.\n",
    "\n",
    "Пример ответа:  0.06"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.008635460765911776\n",
      "[0.1460658 0.0029994]\n"
     ]
    }
   ],
   "source": [
    "Y = games[\"Other_Sales\"]\n",
    "X = games[[\"NA_Sales\", \"JP_Sales\"]]\n",
    "\n",
    "model02 = lm.fit(X, Y)\n",
    "print(model02.intercept_) # ответ 0.01\n",
    "print(model02.coef_)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Задача 4 (из тренировочной версии независимого экзамена)\n",
    "\n",
    "Оцените линейную регрессию, уравнение которой имеет следующий вид:\n",
    "\n",
    "$$\n",
    "\\widehat{\\text{Other_Sales}}_i=\\hat{\\omega}_0+\\hat{\\omega}_1 * \\text{NA_Sales}_i + \\hat{\\omega}_2 * \\text{JP_Sales}_i,\n",
    "$$\n",
    "\n",
    "где нижние индексы обозначают значения соответствующих переменных для наблюдения $i$.\n",
    "\n",
    "В ответ выпишите значение MAE (средней абсолютной ошибки) на обучающей выборке.\n",
    "\n",
    "Пример ответа:  0.06"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE: 0.02063977350506148\n",
      "MAE: 0.03508696969377178\n",
      "R2: 0.4078984126724643\n"
     ]
    }
   ],
   "source": [
    "Y_pred2 = model02.predict(X)\n",
    "\n",
    "print(\"MSE:\", mean_squared_error(Y, Y_pred2)) \n",
    "print(\"MAE:\", mean_absolute_error(Y, Y_pred2)) # ответ 0.04\n",
    "print(\"R2:\", r2_score(Y, Y_pred2)) "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Часть 3: множественная линейная регрессия, деление на обучающую и тестовую выборку"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Импортируем необходимые функции:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LinearRegression"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Линейная модель\n",
    "\n",
    "Создаем объект класса `LinearRegression`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "lm = LinearRegression()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Выбираем столбцы:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "X = games[[\"NA_Sales\", \"JP_Sales\"]]\n",
    "Y = games[\"Global_Sales\"]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Разбиваем данные на обучающую и тестовую выборки:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, Y_train, Y_test = train_test_split(X, Y, test_size=0.2, random_state=123)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Обучаем линейную модель:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_ols = lm.fit(X_train, Y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Оцениваем ее качество:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.9375773989782563\n",
      "0.8934811921779167\n"
     ]
    }
   ],
   "source": [
    "print(model_ols.score(X_train, Y_train))\n",
    "print(model_ols.score(X_test, Y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Полиномиальная модель"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Импортируем необходимую функцию:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import PolynomialFeatures"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Преобразуем исходные данные (многочлен третьей степени – вспомните график в начале):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "poly = PolynomialFeatures(degree = 3)\n",
    "X_train_poly = poly.fit_transform(X_train)\n",
    "X_test_poly = poly.fit_transform(X_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Обучаем модель:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "model_poly = lm.fit(X_train_poly, Y_train)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Проверяем качество:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.950115432784038\n",
      "0.8550552695342691\n"
     ]
    }
   ],
   "source": [
    "print(model_poly.score(X_train_poly, Y_train))\n",
    "print(model_poly.score(X_test_poly, Y_test))"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}