{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Задача 14 \n",
    "\n",
    "Используя данные о школьниках, выявить степень их алкогольной зависимости. В данных, взятых с UCI 'Students' (исходная выборка изъята из UCI, но осталась в других источниках), содержится информация о 30 признаках для каждого школьника, включая социальные и гендерные, а также указана материальная обеспеченность и количество свободного времени. Выбрать на свой взгляд наиболее весомые признаки и предсказать степень употребления алкоголя по выходным или будним по шкале от 0 до 5.  \n",
    "\n",
    "Данные: https://github.com/amanchoudhary/student-alcohol-consumption-prediction"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "from sklearn.preprocessing import OneHotEncoder, scale\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LinearRegression\n",
    "from sklearn.metrics import mean_squared_error, accuracy_score\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "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>school</th>\n",
       "      <th>sex</th>\n",
       "      <th>age</th>\n",
       "      <th>address</th>\n",
       "      <th>famsize</th>\n",
       "      <th>Pstatus</th>\n",
       "      <th>Medu</th>\n",
       "      <th>Fedu</th>\n",
       "      <th>Mjob</th>\n",
       "      <th>Fjob</th>\n",
       "      <th>...</th>\n",
       "      <th>famrel</th>\n",
       "      <th>freetime</th>\n",
       "      <th>goout</th>\n",
       "      <th>Dalc</th>\n",
       "      <th>Walc</th>\n",
       "      <th>health</th>\n",
       "      <th>absences</th>\n",
       "      <th>G1</th>\n",
       "      <th>G2</th>\n",
       "      <th>G3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>GP</td>\n",
       "      <td>F</td>\n",
       "      <td>18</td>\n",
       "      <td>U</td>\n",
       "      <td>GT3</td>\n",
       "      <td>A</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>at_home</td>\n",
       "      <td>teacher</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>GP</td>\n",
       "      <td>F</td>\n",
       "      <td>17</td>\n",
       "      <td>U</td>\n",
       "      <td>GT3</td>\n",
       "      <td>T</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>at_home</td>\n",
       "      <td>other</td>\n",
       "      <td>...</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>9</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>GP</td>\n",
       "      <td>F</td>\n",
       "      <td>15</td>\n",
       "      <td>U</td>\n",
       "      <td>LE3</td>\n",
       "      <td>T</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>at_home</td>\n",
       "      <td>other</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>12</td>\n",
       "      <td>13</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "  school sex  age address famsize Pstatus  Medu  Fedu     Mjob     Fjob ...  \\\n",
       "0     GP   F   18       U     GT3       A     4     4  at_home  teacher ...   \n",
       "1     GP   F   17       U     GT3       T     1     1  at_home    other ...   \n",
       "2     GP   F   15       U     LE3       T     1     1  at_home    other ...   \n",
       "\n",
       "  famrel freetime  goout  Dalc  Walc health absences  G1  G2  G3  \n",
       "0      4        3      4     1     1      3        4   0  11  11  \n",
       "1      5        3      3     1     1      3        2   9  11  11  \n",
       "2      4        3      2     2     3      3        6  12  13  12  \n",
       "\n",
       "[3 rows x 33 columns]"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.read_csv('data.csv')\n",
    "data.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Описание признаков:** \n",
    "\n",
    "| feature | description |\n",
    "|:---:|:---:|\n",
    "| school | student's school (binary: 'GP' - Gabriel Pereira or 'MS' - Mousinho da Silveira) | \n",
    "| sex | student's sex (binary: 'F' - female or 'M' - male) | \n",
    "| age | student's age (numeric: from 15 to 22) | \n",
    "| address | student's home address type (binary: 'U' - urban or 'R' - rural) | \n",
    "| famsize | family size (binary: 'LE3' - less or equal to 3 or 'GT3' - greater than 3) | \n",
    "| Pstatus | parent's cohabitation status (binary: 'T' - living together or 'A' - apart) | \n",
    "| Medu | mother's education (numeric: 0 - none, 1 - primary education (4th grade), 2 – 5th to 9th grade, 3 – secondary education or 4 – higher education) | \n",
    "| Fedu | father's education (numeric: 0 - none, 1 - primary education (4th grade), 2 – 5th to 9th grade, 3 – secondary education or 4 – higher education) | \n",
    "| Mjob | mother's job (nominal: 'teacher', 'health' care related, civil 'services' (e.g. administrative or police), 'at_home' or 'other') | \n",
    "| Fjob | father's job (nominal: 'teacher', 'health' care related, civil 'services' (e.g. administrative or police), 'at_home' or 'other') | \n",
    "| reason | reason to choose this school (nominal: close to 'home', school 'reputation', 'course' preference or 'other') | \n",
    "| guardian | student's guardian (nominal: 'mother', 'father' or 'other'|)\n",
    "| traveltime | home to school travel time (numeric: 1 - <15 min., 2 - 15 to 30 min., 3 - 30 min. to 1 hour, or 4 - >1 hour) | \n",
    "| studytime | weekly study time (numeric: 1 - <2 hours, 2 - 2 to 5 hours, 3 - 5 to 10 hours, or 4 - >10 hours) | \n",
    "| failures | number of past class failures (numeric: n if 1<=n<3, else 4) | \n",
    "| schoolsup | extra educational support (binary: yes or no) | \n",
    "| famsup | family educational support (binary: yes or no) | \n",
    "| paid | extra paid classes within the course subject (Math or Portuguese) (binary: yes or no) | \n",
    "| activities | extra-curricular activities (binary: yes or no) | \n",
    "| nursery | attended nursery school (binary: yes or no) | \n",
    "| higher | wants to take higher education (binary: yes or no) | \n",
    "| internet | Internet access at home (binary: yes or no) | \n",
    "| romantic | with a romantic relationship (binary: yes or no) | \n",
    "| famrel | quality of family relationships (numeric: from 1 - very bad to 5 - excellent) | \n",
    "| freetime | free time after school (numeric: from 1 - very low to 5 - very high) | \n",
    "| goout | going out with friends (numeric: from 1 - very low to 5 - very high) | \n",
    "| Dalc | workday alcohol consumption (numeric: from 1 - very low to 5 - very high) | \n",
    "| Walc | weekend alcohol consumption (numeric: from 1 - very low to 5 - very high) | \n",
    "| health | current health status (numeric: from 1 - very bad to 5 - very good) | \n",
    "| absences | number of school absences (numeric: from 0 to 93) | | "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Подготовим данные: заменим категориальные признаки на бинарные (one-hot encoding), разделим входные признаки и те, которые мы должны предсказать. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "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>GP_school</th>\n",
       "      <th>is_female</th>\n",
       "      <th>age</th>\n",
       "      <th>address_is_rural</th>\n",
       "      <th>big_family</th>\n",
       "      <th>parents_live_apart</th>\n",
       "      <th>Medu</th>\n",
       "      <th>Fedu</th>\n",
       "      <th>Mjob</th>\n",
       "      <th>Fjob</th>\n",
       "      <th>...</th>\n",
       "      <th>famrel</th>\n",
       "      <th>freetime</th>\n",
       "      <th>goout</th>\n",
       "      <th>Dalc</th>\n",
       "      <th>Walc</th>\n",
       "      <th>health</th>\n",
       "      <th>absences</th>\n",
       "      <th>G1</th>\n",
       "      <th>G2</th>\n",
       "      <th>G3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>18</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>at_home</td>\n",
       "      <td>teacher</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>17</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>at_home</td>\n",
       "      <td>other</td>\n",
       "      <td>...</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>9</td>\n",
       "      <td>11</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>at_home</td>\n",
       "      <td>other</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>6</td>\n",
       "      <td>12</td>\n",
       "      <td>13</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 33 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   GP_school  is_female  age  address_is_rural  big_family  \\\n",
       "0          1          1   18                 0           1   \n",
       "1          1          1   17                 0           1   \n",
       "2          1          1   15                 0           0   \n",
       "\n",
       "   parents_live_apart  Medu  Fedu     Mjob     Fjob ... famrel freetime  \\\n",
       "0                   1     4     4  at_home  teacher ...      4        3   \n",
       "1                   0     1     1  at_home    other ...      5        3   \n",
       "2                   0     1     1  at_home    other ...      4        3   \n",
       "\n",
       "   goout  Dalc  Walc  health  absences  G1  G2  G3  \n",
       "0      4     1     1       3         4   0  11  11  \n",
       "1      3     1     1       3         2   9  11  11  \n",
       "2      2     2     3       3         6  12  13  12  \n",
       "\n",
       "[3 rows x 33 columns]"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.replace({'yes': 1, 'no': 0}, inplace=True)\n",
    "data['school'].replace({'GP': 1, 'MS': 0}, inplace=True)\n",
    "data['sex'].replace({'F': 1, 'M': 0}, inplace=True)\n",
    "data['address'].replace({'R': 1, 'U': 0}, inplace=True)\n",
    "data['famsize'].replace({'GT3': 1, 'LE3': 0}, inplace=True)\n",
    "data['Pstatus'].replace({'A': 1, 'T': 0}, inplace=True)\n",
    "data = data.rename(columns={'school': 'GP_school',\n",
    "            'sex': 'is_female',\n",
    "            'address': 'address_is_rural',\n",
    "            'famsize': 'big_family',\n",
    "            'Pstatus': 'parents_live_apart'})\n",
    "data.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "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>GP_school</th>\n",
       "      <th>is_female</th>\n",
       "      <th>age</th>\n",
       "      <th>address_is_rural</th>\n",
       "      <th>big_family</th>\n",
       "      <th>parents_live_apart</th>\n",
       "      <th>Medu</th>\n",
       "      <th>Fedu</th>\n",
       "      <th>traveltime</th>\n",
       "      <th>studytime</th>\n",
       "      <th>...</th>\n",
       "      <th>x0_teacher</th>\n",
       "      <th>x1_at_home</th>\n",
       "      <th>x1_health</th>\n",
       "      <th>x1_services</th>\n",
       "      <th>x1_teacher</th>\n",
       "      <th>x2_course</th>\n",
       "      <th>x2_home</th>\n",
       "      <th>x2_reputation</th>\n",
       "      <th>x3_father</th>\n",
       "      <th>x3_mother</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 42 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   GP_school  is_female   age  address_is_rural  big_family  \\\n",
       "0        1.0        1.0  18.0               0.0         1.0   \n",
       "1        1.0        1.0  17.0               0.0         1.0   \n",
       "2        1.0        1.0  15.0               0.0         0.0   \n",
       "\n",
       "   parents_live_apart  Medu  Fedu  traveltime  studytime    ...      \\\n",
       "0                 1.0   4.0   4.0         2.0        2.0    ...       \n",
       "1                 0.0   1.0   1.0         1.0        2.0    ...       \n",
       "2                 0.0   1.0   1.0         1.0        2.0    ...       \n",
       "\n",
       "   x0_teacher  x1_at_home  x1_health  x1_services  x1_teacher  x2_course  \\\n",
       "0         0.0         0.0        0.0          0.0         1.0        1.0   \n",
       "1         0.0         0.0        0.0          0.0         0.0        1.0   \n",
       "2         0.0         0.0        0.0          0.0         0.0        0.0   \n",
       "\n",
       "   x2_home  x2_reputation  x3_father  x3_mother  \n",
       "0      0.0            0.0        0.0        1.0  \n",
       "1      0.0            0.0        1.0        0.0  \n",
       "2      0.0            0.0        0.0        1.0  \n",
       "\n",
       "[3 rows x 42 columns]"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "enc = OneHotEncoder(sparse=False)\n",
    "categorial = ['Mjob', 'Fjob', 'reason', 'guardian']\n",
    "enc.fit(data[categorial])\n",
    "enc.get_feature_names()\n",
    "encoded = pd.DataFrame(\n",
    "    enc.transform(data[categorial]),\n",
    "    columns=enc.get_feature_names())\n",
    "encoded.drop([f'x{i}_other' for i in range(4)], axis=1, inplace=True)\n",
    "data.drop(categorial, axis=1, inplace=True)\n",
    "data = data.join(encoded)\n",
    "data = data.astype('float64')\n",
    "data.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_Dalc = data['Dalc']\n",
    "y_Walc = data['Walc']\n",
    "data.drop(['Dalc', 'Walc'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Теперь посмотрим, какие признаки кореллируют и уберём лишние."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,8))\n",
    "sns.heatmap(data.corr()) \n",
    "t = plt.title(\"Корреляция признаков\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Итак, из признаков с оценками (`G1, G2, G3`) оставим только `G3`, а `Fedu` и `Medu` усредним."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "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>GP_school</th>\n",
       "      <th>is_female</th>\n",
       "      <th>age</th>\n",
       "      <th>address_is_rural</th>\n",
       "      <th>big_family</th>\n",
       "      <th>parents_live_apart</th>\n",
       "      <th>traveltime</th>\n",
       "      <th>studytime</th>\n",
       "      <th>failures</th>\n",
       "      <th>schoolsup</th>\n",
       "      <th>...</th>\n",
       "      <th>x1_at_home</th>\n",
       "      <th>x1_health</th>\n",
       "      <th>x1_services</th>\n",
       "      <th>x1_teacher</th>\n",
       "      <th>x2_course</th>\n",
       "      <th>x2_home</th>\n",
       "      <th>x2_reputation</th>\n",
       "      <th>x3_father</th>\n",
       "      <th>x3_mother</th>\n",
       "      <th>parents_education</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0.730944</td>\n",
       "      <td>0.833377</td>\n",
       "      <td>1.031695</td>\n",
       "      <td>-0.660182</td>\n",
       "      <td>0.648175</td>\n",
       "      <td>2.666927</td>\n",
       "      <td>0.576718</td>\n",
       "      <td>0.083653</td>\n",
       "      <td>-0.374305</td>\n",
       "      <td>2.923032</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.263045</td>\n",
       "      <td>-0.19168</td>\n",
       "      <td>-0.621894</td>\n",
       "      <td>4.126473</td>\n",
       "      <td>1.130130</td>\n",
       "      <td>-0.545894</td>\n",
       "      <td>-0.53161</td>\n",
       "      <td>-0.555399</td>\n",
       "      <td>0.652973</td>\n",
       "      <td>1.568580</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0.730944</td>\n",
       "      <td>0.833377</td>\n",
       "      <td>0.210137</td>\n",
       "      <td>-0.660182</td>\n",
       "      <td>0.648175</td>\n",
       "      <td>-0.374963</td>\n",
       "      <td>-0.760032</td>\n",
       "      <td>0.083653</td>\n",
       "      <td>-0.374305</td>\n",
       "      <td>-0.342110</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.263045</td>\n",
       "      <td>-0.19168</td>\n",
       "      <td>-0.621894</td>\n",
       "      <td>-0.242338</td>\n",
       "      <td>1.130130</td>\n",
       "      <td>-0.545894</td>\n",
       "      <td>-0.53161</td>\n",
       "      <td>1.800508</td>\n",
       "      <td>-1.531457</td>\n",
       "      <td>-1.392181</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0.730944</td>\n",
       "      <td>0.833377</td>\n",
       "      <td>-1.432980</td>\n",
       "      <td>-0.660182</td>\n",
       "      <td>-1.542792</td>\n",
       "      <td>-0.374963</td>\n",
       "      <td>-0.760032</td>\n",
       "      <td>0.083653</td>\n",
       "      <td>-0.374305</td>\n",
       "      <td>2.923032</td>\n",
       "      <td>...</td>\n",
       "      <td>-0.263045</td>\n",
       "      <td>-0.19168</td>\n",
       "      <td>-0.621894</td>\n",
       "      <td>-0.242338</td>\n",
       "      <td>-0.884854</td>\n",
       "      <td>-0.545894</td>\n",
       "      <td>-0.53161</td>\n",
       "      <td>-0.555399</td>\n",
       "      <td>0.652973</td>\n",
       "      <td>-1.392181</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 37 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   GP_school  is_female       age  address_is_rural  big_family  \\\n",
       "0   0.730944   0.833377  1.031695         -0.660182    0.648175   \n",
       "1   0.730944   0.833377  0.210137         -0.660182    0.648175   \n",
       "2   0.730944   0.833377 -1.432980         -0.660182   -1.542792   \n",
       "\n",
       "   parents_live_apart  traveltime  studytime  failures  schoolsup  \\\n",
       "0            2.666927    0.576718   0.083653 -0.374305   2.923032   \n",
       "1           -0.374963   -0.760032   0.083653 -0.374305  -0.342110   \n",
       "2           -0.374963   -0.760032   0.083653 -0.374305   2.923032   \n",
       "\n",
       "         ...          x1_at_home  x1_health  x1_services  x1_teacher  \\\n",
       "0        ...           -0.263045   -0.19168    -0.621894    4.126473   \n",
       "1        ...           -0.263045   -0.19168    -0.621894   -0.242338   \n",
       "2        ...           -0.263045   -0.19168    -0.621894   -0.242338   \n",
       "\n",
       "   x2_course   x2_home  x2_reputation  x3_father  x3_mother  parents_education  \n",
       "0   1.130130 -0.545894       -0.53161  -0.555399   0.652973           1.568580  \n",
       "1   1.130130 -0.545894       -0.53161   1.800508  -1.531457          -1.392181  \n",
       "2  -0.884854 -0.545894       -0.53161  -0.555399   0.652973          -1.392181  \n",
       "\n",
       "[3 rows x 37 columns]"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['parents_education'] = (data['Fedu'] + data['Medu'])/2\n",
    "data.drop(['Fedu', 'Medu', 'G1', 'G2'], axis=1, inplace=True)\n",
    "data = pd.DataFrame(scale(data), columns=data.columns)\n",
    "data.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(data, y_Dalc, \n",
    "                                                    test_size=0.3,\n",
    "                                                    random_state=22)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Будем рассматривать задачу регрессии (хотя фактически нас просят предсказать 5 классов, но на них есть порядок). В ответе округлим предсказания до ближайшего целого от 1 до 5."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train\n",
      "no round: 0.5958400507521732\n",
      "rounded: 0.7092511013215859\n",
      "accuracy: 0.579295154185022\n",
      "Test\n",
      "no round: 0.8545655788660281\n",
      "rounded: 1.005128205128205\n",
      "accuracy: 0.5128205128205128\n"
     ]
    }
   ],
   "source": [
    "lr = LinearRegression(fit_intercept=True)\n",
    "lr.fit(X_train, y_train)\n",
    "y_pred = lr.predict(X_train)\n",
    "print('Train')\n",
    "print('no round:', mean_squared_error(y_pred, y_train))\n",
    "print('rounded:', mean_squared_error(y_pred.round(), y_train))\n",
    "print('accuracy:', accuracy_score(np.minimum(np.round(y_pred), 5), y_train))\n",
    "\n",
    "print('Test')\n",
    "y_test_pred = lr.predict(X_test)\n",
    "print('no round:', mean_squared_error(y_test_pred, y_test))\n",
    "print('rounded:', mean_squared_error(y_test_pred.round(), y_test))\n",
    "print('accuracy:', accuracy_score(np.minimum(np.round(y_test_pred), 5), y_test))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Выше мы использовали все доступные признаки. Теперь попробуем установить оптимальное число признаков, пользуясь критерием Стьюдента для гипотезы о незначимости каждого из признаков. Отсортируем признаки в порядке возрастания pvalue и будем последовательно добавлять следующие по значимости."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "from statsmodels.regression.linear_model import OLS\n",
    "results = OLS(y_train, X_train, hasconst=True).fit(use_t=True)\n",
    "feature_importance = results.pvalues.to_dict()\n",
    "sorted_features = sorted(feature_importance, key=feature_importance.get)\n",
    "features = len(sorted_features)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "mse_results = []\n",
    "accuracy_results = []\n",
    "mse_results_round = []\n",
    "for k in range(1, features + 1):\n",
    "    lr = LinearRegression(fit_intercept=True)\n",
    "    reduced_features = sorted_features[:k]\n",
    "    X_train_reduced = X_train[reduced_features]\n",
    "    X_test_reduced = X_test[reduced_features]\n",
    "    lr.fit(X_train_reduced, y_train)\n",
    "    \n",
    "    y_pred = lr.predict(X_train_reduced)\n",
    "    y_test_pred = lr.predict(X_test_reduced)\n",
    "    y_pred_rounded = np.minimum(np.round(y_pred), 5)\n",
    "    y_test_pred_rounded = np.minimum(np.round(y_test_pred), 5)\n",
    "    mse_results.append((mean_squared_error(y_pred, y_train),\n",
    "                       mean_squared_error(y_test_pred, y_test)))\n",
    "    mse_results_round.append((mean_squared_error(y_pred_rounded, y_train),\n",
    "                       mean_squared_error(y_test_pred_rounded, y_test)))\n",
    "    accuracy_results.append((accuracy_score(y_pred_rounded, y_train),\n",
    "                       accuracy_score(y_test_pred_rounded, y_test)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "mse_results_train = np.array(mse_results_round).T[0]\n",
    "mse_results_test = np.array(mse_results_round).T[1]\n",
    "acc_results_train = np.array(accuracy_results).T[0]\n",
    "acc_results_test = np.array(accuracy_results).T[1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Посчитаем дисперсию ошибки при взятии разных подвыборок в качестве тестовой."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "indices = np.arange(data.shape[0])\n",
    "indices_shuffled = []\n",
    "test_size = int(np.round(0.3 * data.shape[0]))\n",
    "partitions_size = 100\n",
    "mse_data = np.zeros((partitions_size, features))\n",
    "accuracy_data = np.zeros((partitions_size, features))\n",
    "mse_test_data = np.zeros((partitions_size, features))\n",
    "accuracy_test_data = np.zeros((partitions_size, features))\n",
    "for i in range(partitions_size):\n",
    "    indices_shuffled = np.random.permutation(indices)\n",
    "    X_train = data.iloc[indices_shuffled[test_size:]] \n",
    "    X_test = data.iloc[indices_shuffled[:test_size]] \n",
    "    y_train = y_Dalc[indices_shuffled[test_size:]]\n",
    "    y_test = y_Dalc[indices_shuffled[:test_size]]\n",
    "    for k in range(1, features + 1): \n",
    "        lr = LinearRegression(fit_intercept=True)\n",
    "        reduced_features = sorted_features[:k]\n",
    "        X_train_reduced = X_train[reduced_features]\n",
    "        X_test_reduced = X_test[reduced_features]\n",
    "        lr.fit(X_train_reduced, y_train)\n",
    "        y_pred = lr.predict(X_train_reduced)\n",
    "        y_test_pred = lr.predict(X_test_reduced)\n",
    "        y_pred_rounded = np.minimum(np.round(y_pred), 5)\n",
    "        y_test_pred_rounded = np.minimum(np.round(y_test_pred), 5)\n",
    "        mse_data[i][k-1] = mean_squared_error(y_pred_rounded, y_train)\n",
    "        accuracy_data[i][k-1] = accuracy_score(y_pred_rounded, y_train)\n",
    "        mse_test_data[i][k-1] = mean_squared_error(y_test_pred_rounded, y_test)\n",
    "        accuracy_test_data[i][k-1] = accuracy_score(y_test_pred_rounded, y_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Посмотрим на среднюю дисперсию при разном количестве признаков (более наглядно покажем на графике)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "MSE\n",
      "train: 0.043711321156705604\n",
      "test: 0.0911881493949942\n",
      "Accuracy\n",
      "train: 0.021003891011671237\n",
      "test: 0.035083109815568746\n"
     ]
    }
   ],
   "source": [
    "mse_std = np.std(mse_data, axis=0)\n",
    "mse_test_std = np.std(mse_test_data, axis=0)\n",
    "acc_std = np.std(accuracy_data, axis=0)\n",
    "acc_test_std = np.std(accuracy_test_data, axis=0)\n",
    "print('MSE')\n",
    "print('train:', mse_std.mean())\n",
    "print('test:', mse_test_std.mean())\n",
    "print('Accuracy')\n",
    "print('train:', acc_std.mean())\n",
    "print('test:', acc_test_std.mean())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Построим график зависимости ошибки от количества признаков, указав на нём и дисперсию тоже."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(12, 6))\n",
    "plt.plot(np.arange(1, features + 1), mse_results_test, label='test')\n",
    "plt.fill_between(np.arange(1, features + 1),\n",
    "                 mse_results_test - mse_test_std,\n",
    "                 mse_results_test + mse_test_std,\n",
    "                 alpha = 0.3, label='test_std')\n",
    "\n",
    "plt.plot(np.arange(1, features + 1), mse_results_train, label='train')\n",
    "plt.fill_between(np.arange(1, features + 1),\n",
    "                 mse_results_train - mse_std,\n",
    "                 mse_results_train + mse_std,\n",
    "                 alpha = 0.3, label='train_std')\n",
    "plt.legend()\n",
    "plt.grid()\n",
    "plt.ylabel('MSE')\n",
    "plt.xlabel('количество признаков')\n",
    "t = plt.title('Зависимость MSE от количества признаков')\n",
    "plt.savefig('graph.jpg')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Таким образом, мы видим, что уже после 8 признаков ошибка на контрольной выборке начинает расти."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Accuracy менее важная для нас метрика, но убедимся, что на ней 8 признаков тоже являются одним из оптимальных выборов:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10, 6))\n",
    "plt.plot(np.arange(1, features + 1), np.array(accuracy_results).T[0],\n",
    "        label='train')\n",
    "plt.plot(np.arange(1, features + 1), np.array(accuracy_results).T[1],\n",
    "        label='test')\n",
    "plt.fill_between(np.arange(1, features + 1),\n",
    "                 acc_results_train - acc_std,\n",
    "                 acc_results_train + acc_std,\n",
    "                 alpha = 0.3, label='test_std')\n",
    "plt.fill_between(np.arange(1, features + 1),\n",
    "                 acc_results_test - acc_test_std,\n",
    "                 acc_results_test + acc_test_std,\n",
    "                 alpha = 0.3, label='train_std')\n",
    "plt.legend()\n",
    "plt.grid()\n",
    "plt.ylabel('accuracy')\n",
    "plt.xlabel('количество признаков')\n",
    "t = plt.title('Зависимость accuracy от количества признаков')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}