Lab_1_Exploratory_analysis.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Lab_1.Exploratory_analysis.ipynb",
      "provenance": [],
      "collapsed_sections": [
        "vDY7mAWAqhla"
      ]
    },
    "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": "3.8.5"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5TKas9-HqhlU"
      },
      "source": [
        "# <center> Майнор \"Интеллектуальный анализ данных\" </center>"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Vhq31ofhz-iq",
        "outputId": "73ca869c-de41-4025-91f2-6a9f0f59ecef"
      },
      "source": [
        "from google.colab import drive\n",
        "drive.mount('/content/drive')"
      ],
      "execution_count": null,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Drive already mounted at /content/drive; to attempt to forcibly remount, call drive.mount(\"/content/drive\", force_remount=True).\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "vDY7mAWAqhla"
      },
      "source": [
        "# <center> Курс \"Введение в анализ данных\" </center>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "t7jXybQRqhlb"
      },
      "source": [
        "# <center> Лабораторная работа №1. Обработка и первичный анализ данных </center>"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LcDqFh-xqhlb"
      },
      "source": [
        "## Данные"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "A76BepXnqhlb"
      },
      "source": [
        "В рамках данной лабораторной работы вам предлагается проанализировать набор данных о студентах двух школ в Португалии.  \n",
        "В файле `students_data.csv` представлена информация о студентах, посещающих два курса - математику (`Math`) и поргутальский язык (`Por`). Некоторые студенты представлены в обоих курсах, некоторые - только в одном. Для каждого студента известны три оценки по курсу: оценка за первое полугодие (`G1`), оценка за второе полугодие (`G2`) и итоговая оценка за год (`G3`)."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9gF3Zg9iqhlc"
      },
      "source": [
        "from collections import Counter\n",
        "import pandas as pd\n",
        "import numpy as np\n",
        "import seaborn as sns\n",
        "import matplotlib.pyplot as plt\n",
        "from sklearn.impute import SimpleImputer\n",
        "from sklearn.ensemble import IsolationForest\n",
        "from scipy import stats\n",
        "# from sklearn.preprocessing import OrdinalEncoder\n",
        "\n",
        "import warnings\n",
        "warnings.filterwarnings('ignore')\n",
        "\n",
        "%matplotlib inline"
      ],
      "execution_count": 708,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "WUB4v_EVqhlc"
      },
      "source": [
        "pd.set_option('display.max_columns', None)\n",
        "pd.set_option('display.max_colwidth', None)"
      ],
      "execution_count": 709,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "8OfNsHLUqhld",
        "outputId": "2d5e0b52-414b-4c1d-ed72-2ebaaf60f5e7"
      },
      "source": [
        "data = pd.read_csv(\"/content/drive/MyDrive/IAD/Lab_1/students_data.csv\")\n",
        "data.shape"
      ],
      "execution_count": 710,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(1044, 36)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 710
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 517
        },
        "id": "DmbkvMQ6qhlf",
        "outputId": "0dcb4b94-a669-4e9f-b456-9c1c6b8d6c92"
      },
      "source": [
        "data.head(15)"
      ],
      "execution_count": 711,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>ID</th>\n",
              "      <th>Subject</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>reason</th>\n",
              "      <th>guardian</th>\n",
              "      <th>traveltime</th>\n",
              "      <th>studytime</th>\n",
              "      <th>failures</th>\n",
              "      <th>schoolsup</th>\n",
              "      <th>famsup</th>\n",
              "      <th>paid</th>\n",
              "      <th>activities</th>\n",
              "      <th>nursery</th>\n",
              "      <th>higher</th>\n",
              "      <th>internet</th>\n",
              "      <th>romantic</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>cheating</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>100097</td>\n",
              "      <td>Por</td>\n",
              "      <td>GP</td>\n",
              "      <td>F</td>\n",
              "      <td>16</td>\n",
              "      <td>U</td>\n",
              "      <td>GT3</td>\n",
              "      <td>A</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>services</td>\n",
              "      <td>other</td>\n",
              "      <td>course</td>\n",
              "      <td>father</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>3.0</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>4.0</td>\n",
              "      <td>5</td>\n",
              "      <td>12</td>\n",
              "      <td>NaN</td>\n",
              "      <td>15</td>\n",
              "      <td>13</td>\n",
              "      <td>14</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>101021</td>\n",
              "      <td>Por</td>\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>3</td>\n",
              "      <td>2</td>\n",
              "      <td>other</td>\n",
              "      <td>other</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>5.0</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>1.0</td>\n",
              "      <td>3.0</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>NaN</td>\n",
              "      <td>17</td>\n",
              "      <td>18</td>\n",
              "      <td>17</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>102965</td>\n",
              "      <td>Por</td>\n",
              "      <td>GP</td>\n",
              "      <td>M</td>\n",
              "      <td>16</td>\n",
              "      <td>U</td>\n",
              "      <td>LE3</td>\n",
              "      <td>T</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>health</td>\n",
              "      <td>services</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>4.0</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>3.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>yes</td>\n",
              "      <td>9</td>\n",
              "      <td>8</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>102989</td>\n",
              "      <td>Por</td>\n",
              "      <td>MS</td>\n",
              "      <td>M</td>\n",
              "      <td>17</td>\n",
              "      <td>U</td>\n",
              "      <td>GT3</td>\n",
              "      <td>T</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>other</td>\n",
              "      <td>services</td>\n",
              "      <td>home</td>\n",
              "      <td>father</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>4.0</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>no</td>\n",
              "      <td>14</td>\n",
              "      <td>15</td>\n",
              "      <td>16</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>103131</td>\n",
              "      <td>Por</td>\n",
              "      <td>GP</td>\n",
              "      <td>F</td>\n",
              "      <td>16</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>home</td>\n",
              "      <td>mother</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>NaN</td>\n",
              "      <td>4.0</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>1.0</td>\n",
              "      <td>4.0</td>\n",
              "      <td>5</td>\n",
              "      <td>2</td>\n",
              "      <td>yes</td>\n",
              "      <td>12</td>\n",
              "      <td>13</td>\n",
              "      <td>13</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>103144</td>\n",
              "      <td>Por</td>\n",
              "      <td>GP</td>\n",
              "      <td>M</td>\n",
              "      <td>18</td>\n",
              "      <td>U</td>\n",
              "      <td>LE3</td>\n",
              "      <td>T</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>services</td>\n",
              "      <td>services</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>3.0</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>4.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>NaN</td>\n",
              "      <td>11</td>\n",
              "      <td>11</td>\n",
              "      <td>12</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>105257</td>\n",
              "      <td>Por</td>\n",
              "      <td>MS</td>\n",
              "      <td>F</td>\n",
              "      <td>18</td>\n",
              "      <td>U</td>\n",
              "      <td>GT3</td>\n",
              "      <td>T</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>other</td>\n",
              "      <td>other</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>5</td>\n",
              "      <td>6</td>\n",
              "      <td>NaN</td>\n",
              "      <td>11</td>\n",
              "      <td>12</td>\n",
              "      <td>9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>107639</td>\n",
              "      <td>Por</td>\n",
              "      <td>MS</td>\n",
              "      <td>F</td>\n",
              "      <td>15</td>\n",
              "      <td>R</td>\n",
              "      <td>GT3</td>\n",
              "      <td>T</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>teacher</td>\n",
              "      <td>other</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>1.0</td>\n",
              "      <td>5</td>\n",
              "      <td>1</td>\n",
              "      <td>3.0</td>\n",
              "      <td>5.0</td>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>13</td>\n",
              "      <td>14</td>\n",
              "      <td>14</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>107914</td>\n",
              "      <td>Math</td>\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>3</td>\n",
              "      <td>2</td>\n",
              "      <td>services</td>\n",
              "      <td>other</td>\n",
              "      <td>reputation</td>\n",
              "      <td>mother</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>4.0</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>1.0</td>\n",
              "      <td>1.0</td>\n",
              "      <td>5</td>\n",
              "      <td>10</td>\n",
              "      <td>no</td>\n",
              "      <td>7</td>\n",
              "      <td>6</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>108089</td>\n",
              "      <td>Por</td>\n",
              "      <td>MS</td>\n",
              "      <td>M</td>\n",
              "      <td>16</td>\n",
              "      <td>R</td>\n",
              "      <td>GT3</td>\n",
              "      <td>T</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>other</td>\n",
              "      <td>health</td>\n",
              "      <td>other</td>\n",
              "      <td>mother</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>3.0</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>1.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>5</td>\n",
              "      <td>4</td>\n",
              "      <td>NaN</td>\n",
              "      <td>9</td>\n",
              "      <td>10</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>108254</td>\n",
              "      <td>Por</td>\n",
              "      <td>GP</td>\n",
              "      <td>M</td>\n",
              "      <td>15</td>\n",
              "      <td>U</td>\n",
              "      <td>LE3</td>\n",
              "      <td>T</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>health</td>\n",
              "      <td>services</td>\n",
              "      <td>course</td>\n",
              "      <td>father</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>4.0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>1.0</td>\n",
              "      <td>3.0</td>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>NaN</td>\n",
              "      <td>12</td>\n",
              "      <td>13</td>\n",
              "      <td>12</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>108296</td>\n",
              "      <td>Por</td>\n",
              "      <td>GP</td>\n",
              "      <td>M</td>\n",
              "      <td>18</td>\n",
              "      <td>U</td>\n",
              "      <td>LE3</td>\n",
              "      <td>T</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>services</td>\n",
              "      <td>other</td>\n",
              "      <td>home</td>\n",
              "      <td>mother</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>4.0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>1.0</td>\n",
              "      <td>3.0</td>\n",
              "      <td>5</td>\n",
              "      <td>6</td>\n",
              "      <td>no</td>\n",
              "      <td>16</td>\n",
              "      <td>16</td>\n",
              "      <td>17</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>108753</td>\n",
              "      <td>Math</td>\n",
              "      <td>MS</td>\n",
              "      <td>F</td>\n",
              "      <td>18</td>\n",
              "      <td>R</td>\n",
              "      <td>GT3</td>\n",
              "      <td>T</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>teacher</td>\n",
              "      <td>at-home</td>\n",
              "      <td>reputation</td>\n",
              "      <td>mother</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>4.0</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>2.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>5</td>\n",
              "      <td>7</td>\n",
              "      <td>NaN</td>\n",
              "      <td>6</td>\n",
              "      <td>5</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>108825</td>\n",
              "      <td>Por</td>\n",
              "      <td>GP</td>\n",
              "      <td>M</td>\n",
              "      <td>15</td>\n",
              "      <td>U</td>\n",
              "      <td>GT3</td>\n",
              "      <td>T</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>teacher</td>\n",
              "      <td>other</td>\n",
              "      <td>home</td>\n",
              "      <td>mother</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>4.0</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>2.0</td>\n",
              "      <td>3.0</td>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>yes</td>\n",
              "      <td>11</td>\n",
              "      <td>11</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>114462</td>\n",
              "      <td>Math</td>\n",
              "      <td>GP</td>\n",
              "      <td>F</td>\n",
              "      <td>16</td>\n",
              "      <td>U</td>\n",
              "      <td>LE3</td>\n",
              "      <td>T</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>at_home</td>\n",
              "      <td>other</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>4.0</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>1.0</td>\n",
              "      <td>2.0</td>\n",
              "      <td>2</td>\n",
              "      <td>4</td>\n",
              "      <td>yes</td>\n",
              "      <td>8</td>\n",
              "      <td>7</td>\n",
              "      <td>6</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "        ID Subject school sex  age address famsize Pstatus Medu Fedu  \\\n",
              "0   100097     Por     GP   F   16       U     GT3       A    3    4   \n",
              "1   101021     Por     GP   F   17       U     GT3       T    3    2   \n",
              "2   102965     Por     GP   M   16       U     LE3       T    1    2   \n",
              "3   102989     Por     MS   M   17       U     GT3       T    2    3   \n",
              "4   103131     Por     GP   F   16       U     GT3       T    1    1   \n",
              "5   103144     Por     GP   M   18       U     LE3       T    3    1   \n",
              "6   105257     Por     MS   F   18       U     GT3       T    1    1   \n",
              "7   107639     Por     MS   F   15       R     GT3       T    4    4   \n",
              "8   107914    Math     GP   F   15       U     LE3       T    3    2   \n",
              "9   108089     Por     MS   M   16       R     GT3       T    3    4   \n",
              "10  108254     Por     GP   M   15       U     LE3       T    4    4   \n",
              "11  108296     Por     GP   M   18       U     LE3       T    3    4   \n",
              "12  108753    Math     MS   F   18       R     GT3       T    4    4   \n",
              "13  108825     Por     GP   M   15       U     GT3       T    4    3   \n",
              "14  114462    Math     GP   F   16       U     LE3       T    2    2   \n",
              "\n",
              "        Mjob      Fjob      reason guardian  traveltime  studytime  failures  \\\n",
              "0   services     other      course   father           1          1         0   \n",
              "1      other     other      course   mother           1          2         0   \n",
              "2     health  services      course   mother           2          1         2   \n",
              "3      other  services        home   father           2          2         0   \n",
              "4    at_home     other        home   mother           2          1         0   \n",
              "5   services  services      course   mother           2          1         0   \n",
              "6      other     other      course   mother           2          2         0   \n",
              "7    teacher     other      course   mother           2          1         0   \n",
              "8   services     other  reputation   mother           1          2         0   \n",
              "9      other    health       other   mother           3          2         0   \n",
              "10    health  services      course   father           1          1         0   \n",
              "11  services     other        home   mother           1          2         0   \n",
              "12   teacher   at-home  reputation   mother           3          1         0   \n",
              "13   teacher     other        home   mother           1          2         0   \n",
              "14   at_home     other      course   mother           1          2         0   \n",
              "\n",
              "   schoolsup famsup paid activities nursery higher internet romantic  famrel  \\\n",
              "0         no     no   no         no     yes    yes      yes       no     3.0   \n",
              "1         no     no   no        yes      no    yes      yes       no     5.0   \n",
              "2         no     no   no         no      no    yes      yes       no     4.0   \n",
              "3         no     no   no        yes     yes    yes      yes       no     4.0   \n",
              "4         no    yes   no         no     yes    yes       no      NaN     4.0   \n",
              "5         no     no   no        yes     yes    yes      yes      yes     3.0   \n",
              "6         no     no   no        yes     yes    yes       no       no     1.0   \n",
              "7         no     no   no         no     yes    yes      yes      yes     1.0   \n",
              "8         no    yes  yes         no     yes    yes      yes       no     4.0   \n",
              "9         no     no   no         no      no    yes       no       no     3.0   \n",
              "10        no    yes   no        yes     yes    yes      yes       no     4.0   \n",
              "11        no     no   no        yes     yes    yes      yes      yes     4.0   \n",
              "12        no    yes  yes        yes     yes    yes      yes      yes     4.0   \n",
              "13        no    yes   no        yes     yes    yes      yes       no     4.0   \n",
              "14        no    yes   no         no     yes    yes       no       no     4.0   \n",
              "\n",
              "    freetime  goout  Dalc  Walc  health  absences cheating  G1  G2  G3  \n",
              "0          2      1   1.0   4.0       5        12      NaN  15  13  14  \n",
              "1          3      4   1.0   3.0       3         2      NaN  17  18  17  \n",
              "2          4      5   3.0   5.0       5         0      yes   9   8  10  \n",
              "3          4      3   1.0   1.0       3         4       no  14  15  16  \n",
              "4          3      2   1.0   4.0       5         2      yes  12  13  13  \n",
              "5          3      4   4.0   5.0       4         2      NaN  11  11  12  \n",
              "6          1      1   1.0   1.0       5         6      NaN  11  12   9  \n",
              "7          5      1   3.0   5.0       5         0      NaN  13  14  14  \n",
              "8          4      4   1.0   1.0       5        10       no   7   6   6  \n",
              "9          4      5   1.0   2.0       5         4      NaN   9  10  11  \n",
              "10         3      3   1.0   3.0       5         0      NaN  12  13  12  \n",
              "11         3      3   1.0   3.0       5         6       no  16  16  17  \n",
              "12         4      3   2.0   2.0       5         7      NaN   6   5   6  \n",
              "13         3      3   2.0   3.0       5         0      yes  11  11  11  \n",
              "14         3      4   1.0   2.0       2         4      yes   8   7   6  "
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 711
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "OeQJ3fAZqhlg"
      },
      "source": [
        "### Признаки"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "iKifGCNUqhlg"
      },
      "source": [
        "Данные представлены признаками различных типов: числовыми, категориальными, упорядоченными категориальными."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "ZikyuAPwqhlg"
      },
      "source": [
        "**Описание признаков:**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "wvhg2F4pqhlh",
        "outputId": "ad6b78fb-62cd-43de-dc76-8a867a087934"
      },
      "source": [
        "pd.read_csv('/content/drive/MyDrive/IAD/Lab_1/students_data_features.csv',\n",
        "            delimiter=';',\n",
        "            encoding='windows-1251')"
      ],
      "execution_count": 712,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>Признак</th>\n",
              "      <th>Описание</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>ID</td>\n",
              "      <td>Уникальный номер наблюдения</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Subject</td>\n",
              "      <td>Предмет: 'Math' - математика, 'Por' - португальский язык</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>school</td>\n",
              "      <td>Школа: 'GP' - Gabriel Pereira, 'MS' - Mousinho da Silveira</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>sex</td>\n",
              "      <td>Пол студента: 'F' - женский, 'M' - мужской</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>age</td>\n",
              "      <td>Возраст студента</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>address</td>\n",
              "      <td>Место проживания студента: 'U' - в городе, 'R' - за городом</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>famsize</td>\n",
              "      <td>Число членов семьи: 'LE3' - если &lt;= 3, 'GT3' - если &gt; 3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>Pstatus</td>\n",
              "      <td>Отношения родителей: 'T' - живут вместе, 'A' - живут раздельно</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>Medu</td>\n",
              "      <td>Образование матери: 0 - без образования, 1 - начальная школа, 2 – от 5 до 9 классов, 3 – среднее образование, 4 – высшее образование</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>Fedu</td>\n",
              "      <td>Образование отца: 0 - без образования, 1 - начальная школа, 2 – от 5 до 9 классов, 3 – среднее образование, 4 – высшее образование</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>Mjob</td>\n",
              "      <td>Работа матери: 'teacher' - связана с преподаванием, 'health'  - связана с здравоохранением, 'services' - администрация и полиция, 'at_home' - домашняя занятость, 'other' - другое</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>Fjob</td>\n",
              "      <td>Работа отца: 'teacher' - связана с преподаванием, 'health'  - связана с здравоохранением, 'services' - администрация и полиция, 'at_home' - домашняя занятость, 'other' - другое</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>reason</td>\n",
              "      <td>Причина, по которой студент выбрал данную школу: 'home' - рядом с домом, 'reputation' - репутация школы, 'course' - читаемые курсы, 'other' - другое</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>guardian</td>\n",
              "      <td>Опекун студента (student's guardian): 'mother', 'father' или 'other'</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>traveltime</td>\n",
              "      <td>Время в пути от дома до школы: 1 - &lt; 15 мин, 2 - от 15 до 30 мин, 3 - от 30 мин до 1 часа, 4 - &gt; 1 часа</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>studytime</td>\n",
              "      <td>Количество часов на учебу в неделю: 1 - &lt; 2 часов, 2 - от 2 до 5 часов, 3 - от 5 до 10 часов, 4 - &gt; 10 часов</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>failures</td>\n",
              "      <td>Число завалов студента по предыдущим курсам:  n если 0&lt;=n&lt;3, иначе 3</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17</th>\n",
              "      <td>schoolsup</td>\n",
              "      <td>Дополнительная помощь в учебе: yes или no</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>famsup</td>\n",
              "      <td>Дополнительная помощь в учебе от членов семьи: yes или no</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19</th>\n",
              "      <td>paid</td>\n",
              "      <td>Платные дополнительные занятия по соответствущему курсу ('Math' или 'Por'): yes или no</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>20</th>\n",
              "      <td>activities</td>\n",
              "      <td>Внеучебные активности: yes или no</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>21</th>\n",
              "      <td>nursery</td>\n",
              "      <td>Ходил в детстве в детский сад: yes или no</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>22</th>\n",
              "      <td>higher</td>\n",
              "      <td>Собирается получать высшее образование: yes или no</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23</th>\n",
              "      <td>internet</td>\n",
              "      <td>Есть доступ к всемирной сети из дома: yes или no</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24</th>\n",
              "      <td>romantic</td>\n",
              "      <td>Состоит в романтических отношениях: yes или no</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25</th>\n",
              "      <td>famrel</td>\n",
              "      <td>Отношения в семье: от 1 - очень плохие, до 5 - прекрасные</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>26</th>\n",
              "      <td>freetime</td>\n",
              "      <td>Свободное время после занятий: от 1 - очень мало, до 5 - очень много</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>27</th>\n",
              "      <td>goout</td>\n",
              "      <td>Проводит время с друзьями: от 1 - очень редко/мало, до 5 - очень часто/много</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>28</th>\n",
              "      <td>Dalc</td>\n",
              "      <td>Потребление алкоголя в будни: от 1 - очень мало, до 5 - очень много</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>29</th>\n",
              "      <td>Walc</td>\n",
              "      <td>Потребление алкоголя в выходные: от 1 - очень мало, до 5 - очень много</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>30</th>\n",
              "      <td>health</td>\n",
              "      <td>Состояние здоровья: от 1 - очень плохое, до 5  - очень хорошее</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>31</th>\n",
              "      <td>absences</td>\n",
              "      <td>Число пропусков: сколько раз пропустил занятия в школе</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>32</th>\n",
              "      <td>cheating</td>\n",
              "      <td>Списывает ли на тестах: yes или no</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>33</th>\n",
              "      <td>G1</td>\n",
              "      <td>Оценка за первое полугодние: от 0 до 20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>34</th>\n",
              "      <td>G2</td>\n",
              "      <td>Оценка за второе полугодие: от 0 до 20</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35</th>\n",
              "      <td>G3</td>\n",
              "      <td>Итоговая оценка: от 0 до 20</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       Признак  \\\n",
              "0           ID   \n",
              "1      Subject   \n",
              "2       school   \n",
              "3          sex   \n",
              "4          age   \n",
              "5      address   \n",
              "6      famsize   \n",
              "7      Pstatus   \n",
              "8         Medu   \n",
              "9         Fedu   \n",
              "10        Mjob   \n",
              "11        Fjob   \n",
              "12      reason   \n",
              "13    guardian   \n",
              "14  traveltime   \n",
              "15   studytime   \n",
              "16    failures   \n",
              "17   schoolsup   \n",
              "18      famsup   \n",
              "19        paid   \n",
              "20  activities   \n",
              "21     nursery   \n",
              "22      higher   \n",
              "23    internet   \n",
              "24    romantic   \n",
              "25      famrel   \n",
              "26    freetime   \n",
              "27       goout   \n",
              "28        Dalc   \n",
              "29        Walc   \n",
              "30      health   \n",
              "31    absences   \n",
              "32    cheating   \n",
              "33          G1   \n",
              "34          G2   \n",
              "35          G3   \n",
              "\n",
              "                                                                                                                                                                              Описание  \n",
              "0                                                                                                                                                          Уникальный номер наблюдения  \n",
              "1                                                                                                                             Предмет: 'Math' - математика, 'Por' - португальский язык  \n",
              "2                                                                                                                           Школа: 'GP' - Gabriel Pereira, 'MS' - Mousinho da Silveira  \n",
              "3                                                                                                                                           Пол студента: 'F' - женский, 'M' - мужской  \n",
              "4                                                                                                                                                                     Возраст студента  \n",
              "5                                                                                                                          Место проживания студента: 'U' - в городе, 'R' - за городом  \n",
              "6                                                                                                                              Число членов семьи: 'LE3' - если <= 3, 'GT3' - если > 3  \n",
              "7                                                                                                                       Отношения родителей: 'T' - живут вместе, 'A' - живут раздельно  \n",
              "8                                                 Образование матери: 0 - без образования, 1 - начальная школа, 2 – от 5 до 9 классов, 3 – среднее образование, 4 – высшее образование  \n",
              "9                                                   Образование отца: 0 - без образования, 1 - начальная школа, 2 – от 5 до 9 классов, 3 – среднее образование, 4 – высшее образование  \n",
              "10  Работа матери: 'teacher' - связана с преподаванием, 'health'  - связана с здравоохранением, 'services' - администрация и полиция, 'at_home' - домашняя занятость, 'other' - другое  \n",
              "11    Работа отца: 'teacher' - связана с преподаванием, 'health'  - связана с здравоохранением, 'services' - администрация и полиция, 'at_home' - домашняя занятость, 'other' - другое  \n",
              "12                                Причина, по которой студент выбрал данную школу: 'home' - рядом с домом, 'reputation' - репутация школы, 'course' - читаемые курсы, 'other' - другое  \n",
              "13                                                                                                                Опекун студента (student's guardian): 'mother', 'father' или 'other'  \n",
              "14                                                                             Время в пути от дома до школы: 1 - < 15 мин, 2 - от 15 до 30 мин, 3 - от 30 мин до 1 часа, 4 - > 1 часа  \n",
              "15                                                                        Количество часов на учебу в неделю: 1 - < 2 часов, 2 - от 2 до 5 часов, 3 - от 5 до 10 часов, 4 - > 10 часов  \n",
              "16                                                                                                                Число завалов студента по предыдущим курсам:  n если 0<=n<3, иначе 3  \n",
              "17                                                                                                                                           Дополнительная помощь в учебе: yes или no  \n",
              "18                                                                                                                           Дополнительная помощь в учебе от членов семьи: yes или no  \n",
              "19                                                                                              Платные дополнительные занятия по соответствущему курсу ('Math' или 'Por'): yes или no  \n",
              "20                                                                                                                                                   Внеучебные активности: yes или no  \n",
              "21                                                                                                                                           Ходил в детстве в детский сад: yes или no  \n",
              "22                                                                                                                                  Собирается получать высшее образование: yes или no  \n",
              "23                                                                                                                                    Есть доступ к всемирной сети из дома: yes или no  \n",
              "24                                                                                                                                      Состоит в романтических отношениях: yes или no  \n",
              "25                                                                                                                           Отношения в семье: от 1 - очень плохие, до 5 - прекрасные  \n",
              "26                                                                                                                Свободное время после занятий: от 1 - очень мало, до 5 - очень много  \n",
              "27                                                                                                        Проводит время с друзьями: от 1 - очень редко/мало, до 5 - очень часто/много  \n",
              "28                                                                                                                 Потребление алкоголя в будни: от 1 - очень мало, до 5 - очень много  \n",
              "29                                                                                                              Потребление алкоголя в выходные: от 1 - очень мало, до 5 - очень много  \n",
              "30                                                                                                                      Состояние здоровья: от 1 - очень плохое, до 5  - очень хорошее  \n",
              "31                                                                                                                              Число пропусков: сколько раз пропустил занятия в школе  \n",
              "32                                                                                                                                                  Списывает ли на тестах: yes или no  \n",
              "33                                                                                                                                             Оценка за первое полугодние: от 0 до 20  \n",
              "34                                                                                                                                              Оценка за второе полугодие: от 0 до 20  \n",
              "35                                                                                                                                                         Итоговая оценка: от 0 до 20  "
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 712
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_EwDMtVsqhlh"
      },
      "source": [
        "<br>  \n",
        "\n",
        "## Задания"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "tMPQ7vaYqhlh"
      },
      "source": [
        "**NB:** Обратите внимание, что часть студентов посещают оба курса и представлены в данных два раза. Из-за этого некоторые результаты анализа могут быть искажены. Имеет смысл выявить таких студентов заранее ([Задание 11](#task11)) и учитывать эту информацию при анализе."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rsG13PWSqhli"
      },
      "source": [
        "### Задание 1  \n",
        "  \n",
        " * Для каждого признака определите, к какому типу он относится.   \n",
        " * Проверьте значения признаков на наличие опечаток и ошибок, исправьте их.<br>\n",
        "   __Note:__ Исправлять ошибки нужно программно, модифицировать руками исходный файл нельзя.\n",
        " * При необходимости, приведите признак к определенному вами типу."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "R3paPM9Yqhli",
        "outputId": "28e85bb2-d344-41ca-ba40-0c1535850437"
      },
      "source": [
        "# Исходные типы признаков\n",
        "data.dtypes"
      ],
      "execution_count": 713,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "ID              int64\n",
              "Subject        object\n",
              "school         object\n",
              "sex            object\n",
              "age             int64\n",
              "address        object\n",
              "famsize        object\n",
              "Pstatus        object\n",
              "Medu           object\n",
              "Fedu           object\n",
              "Mjob           object\n",
              "Fjob           object\n",
              "reason         object\n",
              "guardian       object\n",
              "traveltime      int64\n",
              "studytime       int64\n",
              "failures        int64\n",
              "schoolsup      object\n",
              "famsup         object\n",
              "paid           object\n",
              "activities     object\n",
              "nursery        object\n",
              "higher         object\n",
              "internet       object\n",
              "romantic       object\n",
              "famrel        float64\n",
              "freetime        int64\n",
              "goout           int64\n",
              "Dalc          float64\n",
              "Walc          float64\n",
              "health          int64\n",
              "absences        int64\n",
              "cheating       object\n",
              "G1              int64\n",
              "G2              int64\n",
              "G3              int64\n",
              "dtype: object"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 713
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "bmAwyvhv07yU",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "7e5627a4-b5ed-40a9-a1b5-ec429ccbf163"
      },
      "source": [
        "for i in data.columns:\n",
        "  print(f\"Column '{i}' values: {data.loc[:, i].unique()}\")"
      ],
      "execution_count": 714,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Column 'ID' values: [100097 101021 102965 ... 998425 999143 999194]\n",
            "Column 'Subject' values: ['Por' 'Math']\n",
            "Column 'school' values: ['GP' 'MS']\n",
            "Column 'sex' values: ['F' 'M' 'm']\n",
            "Column 'age' values: [16 17 18 15 19 21 20 22]\n",
            "Column 'address' values: ['U' 'R']\n",
            "Column 'famsize' values: ['GT3' 'LE3']\n",
            "Column 'Pstatus' values: ['A' 'T' 't']\n",
            "Column 'Medu' values: ['3' '1' '2' '4' '0' 'o']\n",
            "Column 'Fedu' values: ['4' '2' '3' '1' 'o' '0']\n",
            "Column 'Mjob' values: ['services' 'other' 'health' 'at_home' 'teacher' 'at-home']\n",
            "Column 'Fjob' values: ['other' 'services' 'health' 'at-home' 'teacher' 'at_home']\n",
            "Column 'reason' values: ['course' 'home' 'reputation' 'other']\n",
            "Column 'guardian' values: ['father' 'mother' 'other' 'futher']\n",
            "Column 'traveltime' values: [1 2 3 4]\n",
            "Column 'studytime' values: [1 2 4 3]\n",
            "Column 'failures' values: [0 2 1 3]\n",
            "Column 'schoolsup' values: ['no' 'yes']\n",
            "Column 'famsup' values: ['no' 'yes']\n",
            "Column 'paid' values: ['no' 'yes']\n",
            "Column 'activities' values: ['no' 'yes']\n",
            "Column 'nursery' values: ['yes' 'no']\n",
            "Column 'higher' values: ['yes' 'no']\n",
            "Column 'internet' values: ['yes' 'no']\n",
            "Column 'romantic' values: ['no' nan 'yes']\n",
            "Column 'famrel' values: [ 3.  5.  4.  1.  2. nan]\n",
            "Column 'freetime' values: [2 3 4 1 5]\n",
            "Column 'goout' values: [1 4 5 3 2]\n",
            "Column 'Dalc' values: [ 1.  3.  4.  2.  5. nan]\n",
            "Column 'Walc' values: [ 4.  3.  5.  1.  2. nan]\n",
            "Column 'health' values: [5 3 4 2 1]\n",
            "Column 'absences' values: [12  2  0  4  6 10  7 20 14 11 24  8 16 21  1 18  5 22  9  3 15 26 25 54\n",
            " 56 40 19 13 38 30 17 28 32 23 75]\n",
            "Column 'cheating' values: [nan 'yes' 'no']\n",
            "Column 'G1' values: [15 17  9 14 12 11 13  7 16  6  8 10 18  0  5  3  4 19]\n",
            "Column 'G2' values: [13 18  8 15 11 12 14  6 10 16  5  7  9 17  0  4 19]\n",
            "Column 'G3' values: [14 17 10 16 13 12  9  6 11  8 18 15  0  5  7 19  4  1 20]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "7GVoFBAG2BAy"
      },
      "source": [
        "data = data.replace(\n",
        "    {'sex': {'m': 'M'},\n",
        "     'Pstatus': {'t': 'T'},\n",
        "     'Medu': {'o': '0'},\n",
        "     'Fedu': {'o': '0'},\n",
        "     'Mjob': {'at-home': 'at_home'},\n",
        "     'Fjob': {'at-home': 'at_home'},\n",
        "     'guardian': {'futher': 'father'}\n",
        "    }\n",
        ")"
      ],
      "execution_count": 715,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6_0g9LBN6tqh"
      },
      "source": [
        "data[['Medu', 'Fedu']] = data[['Medu', 'Fedu']].applymap(int)"
      ],
      "execution_count": 716,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "I2Dgk0UQ5I0Y",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "dbd28b97-496e-4e66-9265-f52e0898f93f"
      },
      "source": [
        "for i in data.columns:\n",
        "  print(f\"Column '{i}' values: {data.loc[:, i].unique()}\")"
      ],
      "execution_count": 717,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Column 'ID' values: [100097 101021 102965 ... 998425 999143 999194]\n",
            "Column 'Subject' values: ['Por' 'Math']\n",
            "Column 'school' values: ['GP' 'MS']\n",
            "Column 'sex' values: ['F' 'M']\n",
            "Column 'age' values: [16 17 18 15 19 21 20 22]\n",
            "Column 'address' values: ['U' 'R']\n",
            "Column 'famsize' values: ['GT3' 'LE3']\n",
            "Column 'Pstatus' values: ['A' 'T']\n",
            "Column 'Medu' values: [3 1 2 4 0]\n",
            "Column 'Fedu' values: [4 2 3 1 0]\n",
            "Column 'Mjob' values: ['services' 'other' 'health' 'at_home' 'teacher']\n",
            "Column 'Fjob' values: ['other' 'services' 'health' 'at_home' 'teacher']\n",
            "Column 'reason' values: ['course' 'home' 'reputation' 'other']\n",
            "Column 'guardian' values: ['father' 'mother' 'other']\n",
            "Column 'traveltime' values: [1 2 3 4]\n",
            "Column 'studytime' values: [1 2 4 3]\n",
            "Column 'failures' values: [0 2 1 3]\n",
            "Column 'schoolsup' values: ['no' 'yes']\n",
            "Column 'famsup' values: ['no' 'yes']\n",
            "Column 'paid' values: ['no' 'yes']\n",
            "Column 'activities' values: ['no' 'yes']\n",
            "Column 'nursery' values: ['yes' 'no']\n",
            "Column 'higher' values: ['yes' 'no']\n",
            "Column 'internet' values: ['yes' 'no']\n",
            "Column 'romantic' values: ['no' nan 'yes']\n",
            "Column 'famrel' values: [ 3.  5.  4.  1.  2. nan]\n",
            "Column 'freetime' values: [2 3 4 1 5]\n",
            "Column 'goout' values: [1 4 5 3 2]\n",
            "Column 'Dalc' values: [ 1.  3.  4.  2.  5. nan]\n",
            "Column 'Walc' values: [ 4.  3.  5.  1.  2. nan]\n",
            "Column 'health' values: [5 3 4 2 1]\n",
            "Column 'absences' values: [12  2  0  4  6 10  7 20 14 11 24  8 16 21  1 18  5 22  9  3 15 26 25 54\n",
            " 56 40 19 13 38 30 17 28 32 23 75]\n",
            "Column 'cheating' values: [nan 'yes' 'no']\n",
            "Column 'G1' values: [15 17  9 14 12 11 13  7 16  6  8 10 18  0  5  3  4 19]\n",
            "Column 'G2' values: [13 18  8 15 11 12 14  6 10 16  5  7  9 17  0  4 19]\n",
            "Column 'G3' values: [14 17 10 16 13 12  9  6 11  8 18 15  0  5  7 19  4  1 20]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "w-KbNzEy5QfJ",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "a50c260b-c38b-44aa-c1aa-9260265e6b62"
      },
      "source": [
        "cat_vars = ['Subject', 'school', 'sex', 'address', 'famsize',\n",
        "            'Pstatus', 'Medu', 'Fedu', 'Mjob', 'Fjob', 'reason', 'guardian',\n",
        "            'traveltime', 'studytime', 'schoolsup', 'famsup', 'paid',\n",
        "            'activities', 'nursery', 'higher', 'internet', 'romantic',\n",
        "            'famrel', 'freetime', 'goout', 'Dalc', 'Walc', 'health', 'cheating']\n",
        "\n",
        "for i in cat_vars:\n",
        "  data[i] = data[i].astype('category')\n",
        "\n",
        "data['ID'] = data['ID'].astype('object')\n",
        "\n",
        "data.dtypes"
      ],
      "execution_count": 718,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "ID              object\n",
              "Subject       category\n",
              "school        category\n",
              "sex           category\n",
              "age              int64\n",
              "address       category\n",
              "famsize       category\n",
              "Pstatus       category\n",
              "Medu          category\n",
              "Fedu          category\n",
              "Mjob          category\n",
              "Fjob          category\n",
              "reason        category\n",
              "guardian      category\n",
              "traveltime    category\n",
              "studytime     category\n",
              "failures         int64\n",
              "schoolsup     category\n",
              "famsup        category\n",
              "paid          category\n",
              "activities    category\n",
              "nursery       category\n",
              "higher        category\n",
              "internet      category\n",
              "romantic      category\n",
              "famrel        category\n",
              "freetime      category\n",
              "goout         category\n",
              "Dalc          category\n",
              "Walc          category\n",
              "health        category\n",
              "absences         int64\n",
              "cheating      category\n",
              "G1               int64\n",
              "G2               int64\n",
              "G3               int64\n",
              "dtype: object"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 718
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-ZkiX8aTqhli"
      },
      "source": [
        "### Задание 2  \n",
        "  \n",
        "Проверьте, присутствуют ли в данных пропущенные значения. Сколько? В каких признаках?  \n",
        "По каждому такому признаку предложите и обоснуйте, что делать с пропущенными значениями."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "opYZZaRRqhlj",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "561386c2-c4c0-4dee-e652-f70a6d066874"
      },
      "source": [
        "data.isna().sum()"
      ],
      "execution_count": 719,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "ID              0\n",
              "Subject         0\n",
              "school          0\n",
              "sex             0\n",
              "age             0\n",
              "address         0\n",
              "famsize         0\n",
              "Pstatus         0\n",
              "Medu            0\n",
              "Fedu            0\n",
              "Mjob            0\n",
              "Fjob            0\n",
              "reason          0\n",
              "guardian        0\n",
              "traveltime      0\n",
              "studytime       0\n",
              "failures        0\n",
              "schoolsup       0\n",
              "famsup          0\n",
              "paid            0\n",
              "activities      0\n",
              "nursery         0\n",
              "higher          0\n",
              "internet        0\n",
              "romantic       10\n",
              "famrel         10\n",
              "freetime        0\n",
              "goout           0\n",
              "Dalc           10\n",
              "Walc           10\n",
              "health          0\n",
              "absences        0\n",
              "cheating      700\n",
              "G1              0\n",
              "G2              0\n",
              "G3              0\n",
              "dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 719
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-rbUhr29eqJ2"
      },
      "source": [
        "- Значения признаков **romantic**, **famrel**, **Dalc**, **Walc** пропущены в 10 строках\n",
        "- Значения признака **cheating** пропущены в целых 700 строках!"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NslHK3sSg9LT"
      },
      "source": [
        "data['cheating'] = data['cheating'].fillna('no')"
      ],
      "execution_count": 720,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LGGSR5-ziZRk"
      },
      "source": [
        "- В исходных данных у многих наблюдений признак **cheating** не указан, поэтому человек не может быть обвинён в списывании, если он не был пойман на этом.\n",
        "- Презумпция невиновности в действии! \n",
        "- Удалять этот признак не рекомендуется, потому что он является потенциально важным, хотя пропущен в большом количестве строк. Его удаление может позже вызвать искажение результатов анализа."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gJ0VvuQfj8Gr"
      },
      "source": [
        "data['romantic'] = data['romantic'].fillna('no')"
      ],
      "execution_count": 721,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "aPLopkXllTYL"
      },
      "source": [
        "- Ситуация похожа на ситуацию со списыванием, только пропущенных значений в этот раз гораздо меньше. \n",
        "- Но пропущенные значения всё равно можно заменить на **no**. Если информации об этом нет, значит человек либо специально это скрывает, либо он просто не состоит в отношениях, а поле пропущено по какой-то иной причине (могли просто забыть)."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "g-SbcNHVLIdp"
      },
      "source": [
        "imp = SimpleImputer(strategy=\"most_frequent\")\n",
        "data.iloc[:, :] = imp.fit_transform(data)"
      ],
      "execution_count": 722,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9df0ynMtLulQ"
      },
      "source": [
        "- Признаки **famrel**, **Dalc** и **Walc** также являются категориальными, но с более существенной градацией между категориями.\n",
        "- Вместо того, чтобы удалять строки, мы заполняем эти значения во всего 10 строках вычисленной для каждого признака модой.\n",
        "- Это можно было бы сделать через KNN, но тогда пришлось бы уже на данном этапе перекодировать весь датасет в числовой вид, а нам ещё нужны эти кагории для последющего анализа (как минимум, для удобства)\n",
        "- Мы заполняем значениями именной модой, потому что заполнять константой именно такие значения, зная их описание, достаточно странная идея.\n",
        "- Среднее значение для них было бы крайне странно считать, ибо это, опять же, категориальные переменные."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "VY79RRl-NFil",
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "outputId": "535bd703-6e15-4c3e-bdfa-741bf2c07379"
      },
      "source": [
        "data.isna().sum()"
      ],
      "execution_count": 723,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "ID            0\n",
              "Subject       0\n",
              "school        0\n",
              "sex           0\n",
              "age           0\n",
              "address       0\n",
              "famsize       0\n",
              "Pstatus       0\n",
              "Medu          0\n",
              "Fedu          0\n",
              "Mjob          0\n",
              "Fjob          0\n",
              "reason        0\n",
              "guardian      0\n",
              "traveltime    0\n",
              "studytime     0\n",
              "failures      0\n",
              "schoolsup     0\n",
              "famsup        0\n",
              "paid          0\n",
              "activities    0\n",
              "nursery       0\n",
              "higher        0\n",
              "internet      0\n",
              "romantic      0\n",
              "famrel        0\n",
              "freetime      0\n",
              "goout         0\n",
              "Dalc          0\n",
              "Walc          0\n",
              "health        0\n",
              "absences      0\n",
              "cheating      0\n",
              "G1            0\n",
              "G2            0\n",
              "G3            0\n",
              "dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 723
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "g2KwnE7upsqD"
      },
      "source": [
        "# Приведём famrel из float в int\n",
        "data.loc[:, ['famrel', 'Dalc', 'Walc']] = data.loc[:, ['famrel', 'Dalc', 'Walc']].astype(int)"
      ],
      "execution_count": 724,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "0XHNVImoNIeo"
      },
      "source": [
        "- Всё! Мы смогли обработать пропущенные значения.\n",
        "- Пора перекодировать датафрейм в числа для дальнейшей работы"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "KAh2XFcsqhlj"
      },
      "source": [
        "### Задание 3  \n",
        "  \n",
        "Проанализируйте данные на наличие выбросов.   \n",
        "Сколько наблюдений можно отнести к выбросам по каждому признаку? Сколько всего выбросов в данных?  \n",
        "  \n",
        "**Note:** Выбросы можно не удалять из данных. Это информацию можно учитывать в дальнейшем анализе и удалять выбросы при необходимости. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 608
        },
        "id": "RKhha2j_qhlj",
        "outputId": "ac6f117a-62d2-4b81-88ea-54208d5fdbb2"
      },
      "source": [
        "plt.figure(figsize=(15, 10))\n",
        "sns.boxplot(data=data.drop(columns=['ID']), orient='h')"
      ],
      "execution_count": 725,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fb40d50cc10>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 725
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x720 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9RFFKdc5O3Gg"
      },
      "source": [
        "- Категориальные переменные нам рассматривать нет смысла, потому что там вполне ограниченный и логичный круг значений и о выбросах речи не идёт\n",
        "- Будем рассматривать выбросы числовых значений\n",
        "- Хотя одно из значений возраста на коробке с усами является выбросом, оно вполне возможно в реальной жизни\n",
        "- Хотя много значений в absences выходят за рамки ящика с усами, значения, чуть выходящие за рамки (примерно до 27), распределены достаточно плотно, так что вполне могут быть не выбросами, но стоит посмотреть на эти сэмплы глазами и оценить\n",
        "- Выбросы в G1, G2, G3 выражаются слишком маленькими значениями, но всё же стоит посмотреть своими глазами на эти сэмплы перед их удалением\n",
        "- Кроме того, G3 это итоговая оценка за два семестра, поэтому было бы вполне логично, если бы её значение было бы в районе среднего G1 и G2 у наблюдения\n",
        "- В одном из сэмплов G1 и G2 равны 10, а G3 равно 1, вот это пример того, что очень вряд ли является нормальным наблюдением\n",
        "- Для удобства можем попробовать считать арифметическое среднее и можно сделать поправку +-3 балла от среднего, ибо оно всё-таки не совсем среднее. Это может быть какой-то экзамен или что-то подобное\n",
        "- Но если у человека всё ок было весь год, то почему он должен был завалить экзамен?\n",
        "- Но сразу удалять всё, выходящее за рамки боксплота, не лучшее решение, ибо количество пропусков вполне может коррелировать с оценкой, логично ведь?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "luHBm9rTYw56",
        "outputId": "1e540849-c4e5-4cf5-8d19-98691535a7d3"
      },
      "source": [
        "# Мы рассматриваем числовые признаки absences, G1, G2 и G3\n",
        "z = np.abs(stats.zscore(data.loc[:, ['absences', 'G1', 'G2', 'G3']]))\n",
        "print(z)"
      ],
      "execution_count": 726,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "[[1.21879866 1.26976633 0.53413517 0.68808807]\n",
            " [0.39227478 1.94046507 2.05690172 1.46469775]\n",
            " [0.71448946 0.74232988 0.98863139 0.34739149]\n",
            " ...\n",
            " [0.71448946 1.6051157  1.75234841 1.72356764]\n",
            " [0.07006009 0.74232988 0.68407808 0.0885216 ]\n",
            " [1.54101334 0.26371823 0.07497146 0.0885216 ]]\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "shFK6odRxxO9",
        "outputId": "50ef10e4-374b-4ff7-d69b-ddb822459168"
      },
      "source": [
        "# Задаём threshold для z-score\n",
        "outliers = np.where(z > 3)\n",
        "outliers"
      ],
      "execution_count": 727,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(array([  27,   61,   83,  219,  226,  229,  233,  234,  292,  336,  350,\n",
              "         353,  358,  367,  385,  391,  410,  438,  474,  497,  589,  592,\n",
              "         594,  684,  710,  745,  756,  786,  788,  899,  937,  961,  978,\n",
              "         990, 1026]),\n",
              " array([0, 1, 2, 0, 0, 2, 0, 0, 2, 2, 2, 0, 0, 2, 0, 2, 2, 2, 2, 2, 0, 2,\n",
              "        0, 2, 2, 2, 0, 2, 2, 0, 0, 2, 2, 0, 2]))"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 727
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 217
        },
        "id": "B5T9amY3zjWy",
        "outputId": "35a7a12c-4148-4562-bc1d-05471f7da07f"
      },
      "source": [
        "data.iloc[outliers[0], :].head()"
      ],
      "execution_count": 728,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>ID</th>\n",
              "      <th>Subject</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>reason</th>\n",
              "      <th>guardian</th>\n",
              "      <th>traveltime</th>\n",
              "      <th>studytime</th>\n",
              "      <th>failures</th>\n",
              "      <th>schoolsup</th>\n",
              "      <th>famsup</th>\n",
              "      <th>paid</th>\n",
              "      <th>activities</th>\n",
              "      <th>nursery</th>\n",
              "      <th>higher</th>\n",
              "      <th>internet</th>\n",
              "      <th>romantic</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>cheating</th>\n",
              "      <th>G1</th>\n",
              "      <th>G2</th>\n",
              "      <th>G3</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>27</th>\n",
              "      <td>125309</td>\n",
              "      <td>Math</td>\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>4</td>\n",
              "      <td>4</td>\n",
              "      <td>services</td>\n",
              "      <td>teacher</td>\n",
              "      <td>home</td>\n",
              "      <td>mother</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>24</td>\n",
              "      <td>no</td>\n",
              "      <td>18</td>\n",
              "      <td>18</td>\n",
              "      <td>18</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>61</th>\n",
              "      <td>146260</td>\n",
              "      <td>Por</td>\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>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>no</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>no</td>\n",
              "      <td>0</td>\n",
              "      <td>11</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>83</th>\n",
              "      <td>163217</td>\n",
              "      <td>Por</td>\n",
              "      <td>MS</td>\n",
              "      <td>F</td>\n",
              "      <td>19</td>\n",
              "      <td>R</td>\n",
              "      <td>GT3</td>\n",
              "      <td>A</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>at_home</td>\n",
              "      <td>at_home</td>\n",
              "      <td>course</td>\n",
              "      <td>other</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>3</td>\n",
              "      <td>5</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>yes</td>\n",
              "      <td>8</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>219</th>\n",
              "      <td>278808</td>\n",
              "      <td>Por</td>\n",
              "      <td>GP</td>\n",
              "      <td>F</td>\n",
              "      <td>15</td>\n",
              "      <td>U</td>\n",
              "      <td>GT3</td>\n",
              "      <td>A</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>services</td>\n",
              "      <td>services</td>\n",
              "      <td>home</td>\n",
              "      <td>mother</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>24</td>\n",
              "      <td>yes</td>\n",
              "      <td>9</td>\n",
              "      <td>8</td>\n",
              "      <td>9</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>226</th>\n",
              "      <td>283207</td>\n",
              "      <td>Por</td>\n",
              "      <td>GP</td>\n",
              "      <td>M</td>\n",
              "      <td>18</td>\n",
              "      <td>U</td>\n",
              "      <td>GT3</td>\n",
              "      <td>T</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>other</td>\n",
              "      <td>at_home</td>\n",
              "      <td>course</td>\n",
              "      <td>other</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>26</td>\n",
              "      <td>no</td>\n",
              "      <td>7</td>\n",
              "      <td>8</td>\n",
              "      <td>8</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         ID Subject school sex  age address famsize Pstatus  Medu  Fedu  \\\n",
              "27   125309    Math     GP   F   17       U     GT3       T     4     4   \n",
              "61   146260     Por     GP   F   18       U     GT3       A     4     4   \n",
              "83   163217     Por     MS   F   19       R     GT3       A     1     1   \n",
              "219  278808     Por     GP   F   15       U     GT3       A     3     3   \n",
              "226  283207     Por     GP   M   18       U     GT3       T     2     2   \n",
              "\n",
              "         Mjob      Fjob  reason guardian  traveltime  studytime  failures  \\\n",
              "27   services   teacher    home   mother           2          1         1   \n",
              "61    at_home   teacher  course   mother           2          2         0   \n",
              "83    at_home   at_home  course    other           2          2         3   \n",
              "219  services  services    home   mother           1          2         0   \n",
              "226     other   at_home  course    other           1          1         1   \n",
              "\n",
              "    schoolsup famsup paid activities nursery higher internet romantic  famrel  \\\n",
              "27         no    yes   no         no     yes    yes      yes       no       4   \n",
              "61        yes     no   no         no     yes    yes       no       no       4   \n",
              "83         no    yes   no        yes     yes     no       no      yes       3   \n",
              "219        no     no   no         no      no    yes       no      yes       1   \n",
              "226        no    yes   no        yes      no     no      yes      yes       4   \n",
              "\n",
              "     freetime  goout  Dalc  Walc  health  absences cheating  G1  G2  G3  \n",
              "27          2      4     2     3       2        24       no  18  18  18  \n",
              "61          3      4     1     1       3         4       no   0  11  11  \n",
              "83          5      4     1     4       1         0      yes   8   0   0  \n",
              "219         3      2     2     3       1        24      yes   9   8   9  \n",
              "226         4      3     2     2       1        26       no   7   8   8  "
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 728
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "jfLQXZo4YsEQ",
        "outputId": "22b57670-a3d7-4c2e-9b1a-d2e2383805d9"
      },
      "source": [
        "counter = Counter(outliers[1])\n",
        "\n",
        "print(f'С помощью z-score было найдено {len(outliers[0])} выбросов по 4 признакам, указанным выше.\\n')\n",
        "print(f'Количество выбросов по каждому признаку:\\n')\n",
        "\n",
        "for idx, i in counter.items():\n",
        "  print(f\"{data.loc[:, ['absences', 'G1', 'G2', 'G3']].columns[idx]:8} --> {i}\")"
      ],
      "execution_count": 729,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "С помощью z-score было найдено 35 выбросов по 4 признакам, указанным выше.\n",
            "\n",
            "Количество выбросов по каждому признаку:\n",
            "\n",
            "absences --> 14\n",
            "G1       --> 1\n",
            "G2       --> 20\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vG3wkEJ5TYVT",
        "outputId": "13561cde-e64b-4791-c414-53ba6d40a8ff"
      },
      "source": [
        "Q1 = data.loc[:, ['absences', 'G1', 'G2', 'G3']].quantile(0.25)\n",
        "Q3 = data.loc[:, ['absences', 'G1', 'G2', 'G3']].quantile(0.75)\n",
        "IQR = Q3 - Q1\n",
        "\n",
        "print(f'Q1\\n{Q1}\\n')\n",
        "print(f'Q3\\n{Q3}\\n')\n",
        "print(f'IQR\\n{IQR}\\n')"
      ],
      "execution_count": 730,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Q1\n",
            "absences     0.0\n",
            "G1           9.0\n",
            "G2           9.0\n",
            "G3          10.0\n",
            "Name: 0.25, dtype: float64\n",
            "\n",
            "Q3\n",
            "absences     6.0\n",
            "G1          13.0\n",
            "G2          13.0\n",
            "G3          14.0\n",
            "Name: 0.75, dtype: float64\n",
            "\n",
            "IQR\n",
            "absences    6.0\n",
            "G1          4.0\n",
            "G2          4.0\n",
            "G3          4.0\n",
            "dtype: float64\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "8Ax51fuLUKl1"
      },
      "source": [
        "iqr = ((data < (Q1 - 1.5 * IQR)) |(data > (Q3 + 1.5 * IQR)))"
      ],
      "execution_count": 731,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "WAXIgfNJVqxJ",
        "outputId": "5a2e3b28-b284-4260-8369-a268274303c9"
      },
      "source": [
        "print(f'С помощью IQR (тот же box plot) было найдено {iqr.sum(axis=0).sum()} выбросов по 4 признакам, указанным выше.\\n')\n",
        "print(f'Количество выбросов по каждому признаку:\\n')\n",
        "\n",
        "for i in ['absences', 'G1', 'G2', 'G3']:\n",
        "  print(f\"{i:8} --> {iqr.sum(axis=0)[i]}\")"
      ],
      "execution_count": 732,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "С помощью IQR (тот же box plot) было найдено 129 выбросов по 4 признакам, указанным выше.\n",
            "\n",
            "Количество выбросов по каждому признаку:\n",
            "\n",
            "absences --> 54\n",
            "G1       --> 1\n",
            "G2       --> 20\n",
            "G3       --> 54\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jBWOqJbCW3Gi"
      },
      "source": [
        "clean_data_idx = data[~iqr.any(axis=1)]['ID']"
      ],
      "execution_count": 733,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qchJWCl5cg1M"
      },
      "source": [
        "clean_data = data[data['ID'].isin(clean_data_idx)]"
      ],
      "execution_count": 734,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "KcmCclddcIGT"
      },
      "source": [
        "# Переведём данные в числовой вид\n",
        "enc_data = clean_data.copy()\n",
        "for col in data.columns:\n",
        "  if data[col].dtype == pd.Categorical:\n",
        "    enc_data[col], _ = pd.factorize(clean_data[col])"
      ],
      "execution_count": 735,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "YI5m61Q6uG4P",
        "outputId": "26e53bdd-f2c9-4250-8888-4b8bac77b686"
      },
      "source": [
        "cat_vars = ['Subject', 'school', 'sex', 'address', 'famsize',\n",
        "            'Pstatus', 'Medu', 'Fedu', 'Mjob', 'Fjob', 'reason', 'guardian',\n",
        "            'traveltime', 'studytime', 'schoolsup', 'famsup', 'paid',\n",
        "            'activities', 'nursery', 'higher', 'internet', 'romantic',\n",
        "            'famrel', 'freetime', 'goout', 'Dalc', 'Walc', 'health', 'cheating']\n",
        "\n",
        "for i in cat_vars:\n",
        "  clean_data.loc[:, i] = clean_data[i].astype('category')\n",
        "\n",
        "clean_data.loc[:, 'ID'] = clean_data['ID'].astype('object')\n",
        "\n",
        "clean_data.dtypes"
      ],
      "execution_count": 736,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "ID              object\n",
              "Subject       category\n",
              "school        category\n",
              "sex           category\n",
              "age              int64\n",
              "address       category\n",
              "famsize       category\n",
              "Pstatus       category\n",
              "Medu          category\n",
              "Fedu          category\n",
              "Mjob          category\n",
              "Fjob          category\n",
              "reason        category\n",
              "guardian      category\n",
              "traveltime    category\n",
              "studytime     category\n",
              "failures         int64\n",
              "schoolsup     category\n",
              "famsup        category\n",
              "paid          category\n",
              "activities    category\n",
              "nursery       category\n",
              "higher        category\n",
              "internet      category\n",
              "romantic      category\n",
              "famrel        category\n",
              "freetime      category\n",
              "goout         category\n",
              "Dalc          category\n",
              "Walc          category\n",
              "health        category\n",
              "absences         int64\n",
              "cheating      category\n",
              "G1               int64\n",
              "G2               int64\n",
              "G3               int64\n",
              "dtype: object"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 736
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9QtnbmHBs133"
      },
      "source": [
        "### Задание  11 <a id=\"task11\"></a>  \n",
        "  \n",
        "Часть из представленных в данных студентов посещали оба курса - `Math` и `Por`.  \n",
        "* Предложите способ идентификации студентов и определите тех, кто посещал оба курса. Сколько таких студентов?  \n",
        "* Для этих студентов сравните оценки по двум предметам, дайте комментарий.  \n",
        "  \n",
        "**Note:** Столбец `ID` является уникальным идентификатором наблюдения, но не студента. Он будет уникален для каждой строки."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "E_Xkaql5s6XG"
      },
      "source": [
        "**Сразу перейдём к 11 заданию и заранее выясним студентов, посещающих оба курса.**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "HoTCMEeMtumr"
      },
      "source": [
        "students = data.groupby(data.columns[2:-3].tolist()).groups\n",
        "duplicate_students = [students[x] for x in students if len(students[x]) > 1]\n",
        "\n",
        "duplicate_marks = {}\n",
        "\n",
        "for idx, i in enumerate(duplicate_students):\n",
        "  duplicate_marks[idx, data.iloc[i[0], 1]] = data.iloc[i[0], [0, -3, -2, -1]]\n",
        "  duplicate_marks[idx, data.iloc[i[1], 1]] = data.iloc[i[1], [0, -3, -2, -1]]\n",
        "\n",
        "duplicate_idx = pd.MultiIndex.from_tuples(duplicate_marks)"
      ],
      "execution_count": 737,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 197
        },
        "id": "_u5pVmXZs134",
        "outputId": "fe278eae-5beb-4d6e-f5e3-303a0afb2948"
      },
      "source": [
        "mp = pd.DataFrame(index=duplicate_idx, data=duplicate_marks.values(),\n",
        "                  columns=['ID', 'G1', 'G2', 'G3'])\n",
        "mp.index.set_names(['Student', 'Subject'], inplace=True)\n",
        "mp['ID'] = mp['ID'].astype('object')\n",
        "mp.head(4)"
      ],
      "execution_count": 738,
      "outputs": [
        {
          "output_type": "execute_result",
          "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></th>\n",
              "      <th>ID</th>\n",
              "      <th>G1</th>\n",
              "      <th>G2</th>\n",
              "      <th>G3</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Student</th>\n",
              "      <th>Subject</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th rowspan=\"2\" valign=\"top\">0</th>\n",
              "      <th>Math</th>\n",
              "      <td>257390</td>\n",
              "      <td>10</td>\n",
              "      <td>11</td>\n",
              "      <td>11</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Por</th>\n",
              "      <td>746312</td>\n",
              "      <td>10</td>\n",
              "      <td>10</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th rowspan=\"2\" valign=\"top\">1</th>\n",
              "      <th>Math</th>\n",
              "      <td>544841</td>\n",
              "      <td>9</td>\n",
              "      <td>10</td>\n",
              "      <td>10</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Por</th>\n",
              "      <td>566220</td>\n",
              "      <td>13</td>\n",
              "      <td>12</td>\n",
              "      <td>12</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                     ID  G1  G2  G3\n",
              "Student Subject                    \n",
              "0       Math     257390  10  11  11\n",
              "        Por      746312  10  10  10\n",
              "1       Math     544841   9  10  10\n",
              "        Por      566220  13  12  12"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 738
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 137
        },
        "id": "0IkqH4I8bE4i",
        "outputId": "82ef9999-5573-476d-b745-d63f2d85d5fc"
      },
      "source": [
        "mp.groupby('Subject').mean()"
      ],
      "execution_count": 739,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>G1</th>\n",
              "      <th>G2</th>\n",
              "      <th>G3</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Subject</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Math</th>\n",
              "      <td>11.724138</td>\n",
              "      <td>12.034483</td>\n",
              "      <td>12.344828</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Por</th>\n",
              "      <td>12.379310</td>\n",
              "      <td>12.517241</td>\n",
              "      <td>12.586207</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                G1         G2         G3\n",
              "Subject                                 \n",
              "Math     11.724138  12.034483  12.344828\n",
              "Por      12.379310  12.517241  12.586207"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 739
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "3VnY359sBrqi"
      },
      "source": [
        "- Во втором семестре у этих студентов балл в среднем чуть выше, чем в первом семестре.\n",
        "- По итогам учебного года балл оказался чуть выше, чем отдельно по семестрам."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "9koxy5C_C75p",
        "outputId": "1db0ecd8-ae9e-4a4b-968f-51e2a7653f72"
      },
      "source": [
        "mp.groupby('Subject').mean().sum(axis=1)"
      ],
      "execution_count": 740,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Subject\n",
              "Math    36.103448\n",
              "Por     37.482759\n",
              "dtype: float64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 740
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7TMC2Hn5s136"
      },
      "source": [
        "\r\n",
        "\r\n",
        "*   Студенты почти с одинаковым успехом справились с одновременным посещением двух курсов с небольшим перевесом в пользу португальского\r\n",
        "\r\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "TSWptnVVDGM8"
      },
      "source": [
        "math_idx = mp.xs('Math', level=1)['ID']\n",
        "por_idx = mp.xs('Por', level=1)['ID']"
      ],
      "execution_count": 741,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LDgSga_jNE9d"
      },
      "source": [
        "- Отдельно сохраним ID наблюдений студентов, посещающих два курса"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wwHoBUEtqhlj"
      },
      "source": [
        "### Задание 4 \n",
        "  \n",
        "* Посчитайте описательные статистики для числовых и для категориальных признаков. \n",
        "* Сколько студентов посещают каждую из школ?  \n",
        "* Сколько студентов посещают математику, сколько - португальский?  \n",
        "* Студенты какого курса чаще заваливают экзамены?  \n",
        "* Сколько в среднем времени в неделю уходит на учебу по каждому предмету?  \n",
        "* На что чаще обращают внимание при выборе школы?"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 307
        },
        "id": "-nYswrq6qhlk",
        "outputId": "77d2716f-aa8a-4c65-ae17-8b05d0cd32fc"
      },
      "source": [
        "enc_data.describe()"
      ],
      "execution_count": 742,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>ID</th>\n",
              "      <th>Subject</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>reason</th>\n",
              "      <th>guardian</th>\n",
              "      <th>traveltime</th>\n",
              "      <th>studytime</th>\n",
              "      <th>failures</th>\n",
              "      <th>schoolsup</th>\n",
              "      <th>famsup</th>\n",
              "      <th>paid</th>\n",
              "      <th>activities</th>\n",
              "      <th>nursery</th>\n",
              "      <th>higher</th>\n",
              "      <th>internet</th>\n",
              "      <th>romantic</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>cheating</th>\n",
              "      <th>G1</th>\n",
              "      <th>G2</th>\n",
              "      <th>G3</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>count</th>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "      <td>936.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>mean</th>\n",
              "      <td>548661.392094</td>\n",
              "      <td>0.346154</td>\n",
              "      <td>0.270299</td>\n",
              "      <td>0.434829</td>\n",
              "      <td>16.662393</td>\n",
              "      <td>0.274573</td>\n",
              "      <td>0.299145</td>\n",
              "      <td>0.888889</td>\n",
              "      <td>2.605769</td>\n",
              "      <td>2.394231</td>\n",
              "      <td>1.594017</td>\n",
              "      <td>0.780983</td>\n",
              "      <td>1.035256</td>\n",
              "      <td>0.819444</td>\n",
              "      <td>1.519231</td>\n",
              "      <td>1.994658</td>\n",
              "      <td>0.214744</td>\n",
              "      <td>0.119658</td>\n",
              "      <td>0.616453</td>\n",
              "      <td>0.211538</td>\n",
              "      <td>0.496795</td>\n",
              "      <td>0.197650</td>\n",
              "      <td>0.073718</td>\n",
              "      <td>0.211538</td>\n",
              "      <td>0.334402</td>\n",
              "      <td>3.942308</td>\n",
              "      <td>3.193376</td>\n",
              "      <td>3.145299</td>\n",
              "      <td>1.470085</td>\n",
              "      <td>2.257479</td>\n",
              "      <td>3.544872</td>\n",
              "      <td>3.644231</td>\n",
              "      <td>0.164530</td>\n",
              "      <td>11.481838</td>\n",
              "      <td>11.697650</td>\n",
              "      <td>12.056624</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>std</th>\n",
              "      <td>264435.046029</td>\n",
              "      <td>0.475997</td>\n",
              "      <td>0.444352</td>\n",
              "      <td>0.496000</td>\n",
              "      <td>1.216340</td>\n",
              "      <td>0.446537</td>\n",
              "      <td>0.458128</td>\n",
              "      <td>0.314438</td>\n",
              "      <td>1.135489</td>\n",
              "      <td>1.102030</td>\n",
              "      <td>1.344319</td>\n",
              "      <td>1.155365</td>\n",
              "      <td>1.041306</td>\n",
              "      <td>0.517590</td>\n",
              "      <td>0.736848</td>\n",
              "      <td>0.843328</td>\n",
              "      <td>0.602878</td>\n",
              "      <td>0.324735</td>\n",
              "      <td>0.486510</td>\n",
              "      <td>0.408618</td>\n",
              "      <td>0.500257</td>\n",
              "      <td>0.398439</td>\n",
              "      <td>0.261451</td>\n",
              "      <td>0.408618</td>\n",
              "      <td>0.472033</td>\n",
              "      <td>0.926661</td>\n",
              "      <td>1.031068</td>\n",
              "      <td>1.131106</td>\n",
              "      <td>0.898699</td>\n",
              "      <td>1.277928</td>\n",
              "      <td>1.425556</td>\n",
              "      <td>3.845482</td>\n",
              "      <td>0.370954</td>\n",
              "      <td>2.874504</td>\n",
              "      <td>2.813154</td>\n",
              "      <td>2.859640</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>min</th>\n",
              "      <td>100097.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>15.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>5.000000</td>\n",
              "      <td>5.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25%</th>\n",
              "      <td>324628.250000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>16.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>4.000000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>10.000000</td>\n",
              "      <td>10.000000</td>\n",
              "      <td>10.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50%</th>\n",
              "      <td>557660.500000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>17.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>4.000000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>4.000000</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>11.000000</td>\n",
              "      <td>12.000000</td>\n",
              "      <td>12.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>75%</th>\n",
              "      <td>777731.500000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>18.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>4.000000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>5.000000</td>\n",
              "      <td>4.000000</td>\n",
              "      <td>4.000000</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>5.000000</td>\n",
              "      <td>6.000000</td>\n",
              "      <td>0.000000</td>\n",
              "      <td>14.000000</td>\n",
              "      <td>14.000000</td>\n",
              "      <td>14.000000</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>max</th>\n",
              "      <td>999194.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>22.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>4.000000</td>\n",
              "      <td>4.000000</td>\n",
              "      <td>4.000000</td>\n",
              "      <td>4.000000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>2.000000</td>\n",
              "      <td>4.000000</td>\n",
              "      <td>4.000000</td>\n",
              "      <td>3.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>5.000000</td>\n",
              "      <td>5.000000</td>\n",
              "      <td>5.000000</td>\n",
              "      <td>5.000000</td>\n",
              "      <td>5.000000</td>\n",
              "      <td>5.000000</td>\n",
              "      <td>21.000000</td>\n",
              "      <td>1.000000</td>\n",
              "      <td>19.000000</td>\n",
              "      <td>19.000000</td>\n",
              "      <td>20.000000</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                  ID     Subject      school         sex         age  \\\n",
              "count     936.000000  936.000000  936.000000  936.000000  936.000000   \n",
              "mean   548661.392094    0.346154    0.270299    0.434829   16.662393   \n",
              "std    264435.046029    0.475997    0.444352    0.496000    1.216340   \n",
              "min    100097.000000    0.000000    0.000000    0.000000   15.000000   \n",
              "25%    324628.250000    0.000000    0.000000    0.000000   16.000000   \n",
              "50%    557660.500000    0.000000    0.000000    0.000000   17.000000   \n",
              "75%    777731.500000    1.000000    1.000000    1.000000   18.000000   \n",
              "max    999194.000000    1.000000    1.000000    1.000000   22.000000   \n",
              "\n",
              "          address     famsize     Pstatus        Medu        Fedu        Mjob  \\\n",
              "count  936.000000  936.000000  936.000000  936.000000  936.000000  936.000000   \n",
              "mean     0.274573    0.299145    0.888889    2.605769    2.394231    1.594017   \n",
              "std      0.446537    0.458128    0.314438    1.135489    1.102030    1.344319   \n",
              "min      0.000000    0.000000    0.000000    0.000000    0.000000    0.000000   \n",
              "25%      0.000000    0.000000    1.000000    2.000000    1.000000    1.000000   \n",
              "50%      0.000000    0.000000    1.000000    3.000000    2.000000    1.000000   \n",
              "75%      1.000000    1.000000    1.000000    4.000000    3.000000    3.000000   \n",
              "max      1.000000    1.000000    1.000000    4.000000    4.000000    4.000000   \n",
              "\n",
              "             Fjob      reason    guardian  traveltime   studytime    failures  \\\n",
              "count  936.000000  936.000000  936.000000  936.000000  936.000000  936.000000   \n",
              "mean     0.780983    1.035256    0.819444    1.519231    1.994658    0.214744   \n",
              "std      1.155365    1.041306    0.517590    0.736848    0.843328    0.602878   \n",
              "min      0.000000    0.000000    0.000000    1.000000    1.000000    0.000000   \n",
              "25%      0.000000    0.000000    1.000000    1.000000    1.000000    0.000000   \n",
              "50%      0.000000    1.000000    1.000000    1.000000    2.000000    0.000000   \n",
              "75%      1.000000    2.000000    1.000000    2.000000    2.000000    0.000000   \n",
              "max      4.000000    3.000000    2.000000    4.000000    4.000000    3.000000   \n",
              "\n",
              "        schoolsup      famsup        paid  activities     nursery      higher  \\\n",
              "count  936.000000  936.000000  936.000000  936.000000  936.000000  936.000000   \n",
              "mean     0.119658    0.616453    0.211538    0.496795    0.197650    0.073718   \n",
              "std      0.324735    0.486510    0.408618    0.500257    0.398439    0.261451   \n",
              "min      0.000000    0.000000    0.000000    0.000000    0.000000    0.000000   \n",
              "25%      0.000000    0.000000    0.000000    0.000000    0.000000    0.000000   \n",
              "50%      0.000000    1.000000    0.000000    0.000000    0.000000    0.000000   \n",
              "75%      0.000000    1.000000    0.000000    1.000000    0.000000    0.000000   \n",
              "max      1.000000    1.000000    1.000000    1.000000    1.000000    1.000000   \n",
              "\n",
              "         internet    romantic      famrel    freetime       goout        Dalc  \\\n",
              "count  936.000000  936.000000  936.000000  936.000000  936.000000  936.000000   \n",
              "mean     0.211538    0.334402    3.942308    3.193376    3.145299    1.470085   \n",
              "std      0.408618    0.472033    0.926661    1.031068    1.131106    0.898699   \n",
              "min      0.000000    0.000000    1.000000    1.000000    1.000000    1.000000   \n",
              "25%      0.000000    0.000000    4.000000    3.000000    2.000000    1.000000   \n",
              "50%      0.000000    0.000000    4.000000    3.000000    3.000000    1.000000   \n",
              "75%      0.000000    1.000000    5.000000    4.000000    4.000000    2.000000   \n",
              "max      1.000000    1.000000    5.000000    5.000000    5.000000    5.000000   \n",
              "\n",
              "             Walc      health    absences    cheating          G1          G2  \\\n",
              "count  936.000000  936.000000  936.000000  936.000000  936.000000  936.000000   \n",
              "mean     2.257479    3.544872    3.644231    0.164530   11.481838   11.697650   \n",
              "std      1.277928    1.425556    3.845482    0.370954    2.874504    2.813154   \n",
              "min      1.000000    1.000000    0.000000    0.000000    3.000000    5.000000   \n",
              "25%      1.000000    3.000000    0.000000    0.000000   10.000000   10.000000   \n",
              "50%      2.000000    4.000000    2.000000    0.000000   11.000000   12.000000   \n",
              "75%      3.000000    5.000000    6.000000    0.000000   14.000000   14.000000   \n",
              "max      5.000000    5.000000   21.000000    1.000000   19.000000   19.000000   \n",
              "\n",
              "               G3  \n",
              "count  936.000000  \n",
              "mean    12.056624  \n",
              "std      2.859640  \n",
              "min      5.000000  \n",
              "25%     10.000000  \n",
              "50%     12.000000  \n",
              "75%     14.000000  \n",
              "max     20.000000  "
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 742
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4qm30Z2Ez7nV",
        "outputId": "60589bc5-00f3-4160-ee1e-3ce5378f3b9f"
      },
      "source": [
        "cat_vars = ['Subject', 'school', 'sex', 'address', 'famsize',\n",
        "            'Pstatus', 'Medu', 'Fedu', 'Mjob', 'Fjob', 'reason', 'guardian',\n",
        "            'traveltime', 'studytime', 'schoolsup', 'famsup', 'paid',\n",
        "            'activities', 'nursery', 'higher', 'internet', 'romantic',\n",
        "            'famrel', 'freetime', 'goout', 'Dalc', 'Walc', 'health', 'cheating']\n",
        "\n",
        "for i in cat_vars:\n",
        "  enc_data.loc[:, i] = enc_data[i].astype('category')\n",
        "\n",
        "enc_data.loc[:, 'ID'] = enc_data['ID'].astype('object')\n",
        "\n",
        "enc_data.dtypes"
      ],
      "execution_count": 743,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "ID              object\n",
              "Subject       category\n",
              "school        category\n",
              "sex           category\n",
              "age              int64\n",
              "address       category\n",
              "famsize       category\n",
              "Pstatus       category\n",
              "Medu          category\n",
              "Fedu          category\n",
              "Mjob          category\n",
              "Fjob          category\n",
              "reason        category\n",
              "guardian      category\n",
              "traveltime    category\n",
              "studytime     category\n",
              "failures         int64\n",
              "schoolsup     category\n",
              "famsup        category\n",
              "paid          category\n",
              "activities    category\n",
              "nursery       category\n",
              "higher        category\n",
              "internet      category\n",
              "romantic      category\n",
              "famrel        category\n",
              "freetime      category\n",
              "goout         category\n",
              "Dalc          category\n",
              "Walc          category\n",
              "health        category\n",
              "absences         int64\n",
              "cheating      category\n",
              "G1               int64\n",
              "G2               int64\n",
              "G3               int64\n",
              "dtype: object"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 743
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "_PksN3agVFVm"
      },
      "source": [
        "- Дальнейший анализ будем проводить, не учитывая часть студентов, посещающих два курса\n",
        "- Так как у них почти одна и та же информация (кроме предмета и оценок), то можем произвольно выбирать, что оставляем, а что убираем, если вопрос не касается признаков в скобках\n",
        "- **Проведём анализ, удалив выбросы, обнаруженные с помощью IQR**"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "4ArGG2SAPtqI",
        "outputId": "7edb3916-f3df-435a-aa95-da74f88e0a65"
      },
      "source": [
        "# Количество студентов, посещающих каждую из школ\n",
        "clean_data[~data['ID'].isin(math_idx)].value_counts(subset='school')"
      ],
      "execution_count": 744,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "school\n",
              "GP    658\n",
              "MS    250\n",
              "dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 744
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "XFcJkPDDSp5f",
        "outputId": "f02173fe-25e3-4049-b7ad-455d9f7ab2c1"
      },
      "source": [
        "# Количество студентов, посещающих математику и португальский\n",
        "clean_data.value_counts(subset='Subject')"
      ],
      "execution_count": 745,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Subject\n",
              "Por     612\n",
              "Math    324\n",
              "dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 745
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 137
        },
        "id": "Q21L0GLUS7ac",
        "outputId": "643da38d-b532-4565-f6e5-34d35eea2fc7"
      },
      "source": [
        "# Студенты, посещающие математику, в среднем чаще заваливают другие курсы\n",
        "# (если брать в учёт среднее количество заваленных курсов)\n",
        "clean_data.groupby('Subject')[['failures']].mean()"
      ],
      "execution_count": 746,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>failures</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Subject</th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Math</th>\n",
              "      <td>0.237654</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Por</th>\n",
              "      <td>0.202614</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         failures\n",
              "Subject          \n",
              "Math     0.237654\n",
              "Por      0.202614"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 746
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 137
        },
        "id": "iEWUDXe-tqUo",
        "outputId": "f1283839-9041-44e1-9c50-c9a2fd57856f"
      },
      "source": [
        "# Студенты, посещающие португальский, заваливают больше курсов\n",
        "# (если рассматривать абсолютное количество заваленных курсов)\n",
        "data.groupby('Subject')[['failures']].count()"
      ],
      "execution_count": 747,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>failures</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Subject</th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>Math</th>\n",
              "      <td>395</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>Por</th>\n",
              "      <td>649</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "         failures\n",
              "Subject          \n",
              "Math          395\n",
              "Por           649"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 747
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "gFtiueDHVHO1",
        "outputId": "72e36834-f39b-4b67-8bd2-ee706c82eb45"
      },
      "source": [
        "# На математика уходит в среднем чуть больше времени, хотя результаты близки\n",
        "# и на оба этих предмета ученики тратят обычно от 2 до 5 часов\n",
        "# (округлим 1.95 до 2 ради наглядности)\n",
        "clean_data['studytime'] = clean_data['studytime'].astype(int)\n",
        "clean_data.groupby('Subject')['studytime'].mean()"
      ],
      "execution_count": 748,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Subject\n",
              "Math    2.074074\n",
              "Por     1.952614\n",
              "Name: studytime, dtype: float64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 748
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FrEc8uLjWMxB",
        "outputId": "236a4048-d030-48f0-b9ed-4e4c8696051e"
      },
      "source": [
        "# При выборе школы чаще всего обращают внимание на читаемые курсы\n",
        "clean_data['reason'].value_counts()"
      ],
      "execution_count": 749,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "course        392\n",
              "reputation    227\n",
              "home          218\n",
              "other          99\n",
              "Name: reason, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 749
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "5xxmqfUgqhlk"
      },
      "source": [
        "### Задание 5  \n",
        "  \n",
        "Оцените распределение оценок (отдельно для `G1`, `G2` и `G3`). Сравните распределения для разных предметов - математики и португальского языка. Сделайте выводы.  \n",
        "  \n",
        "**Tip:** Для выполнения задания пригодятся графики плотности, гистограммы и боксплоты."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "wXCBnoGsZ7Ud"
      },
      "source": [
        "- Будем анализировать данные с удалёнными после IQR сэмплами"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 385
        },
        "id": "A4euQGOHqhlk",
        "outputId": "3772411c-4f2d-4872-c864-986c753b7972"
      },
      "source": [
        "sns.catplot(data=clean_data[['Subject', 'G1', 'G2', 'G3']], kind='box', col='Subject')"
      ],
      "execution_count": 750,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<seaborn.axisgrid.FacetGrid at 0x7fb40d224f10>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 750
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "nas6kELJYLdE"
      },
      "source": [
        "- В среднем, диапазон итоговых оценок у португальского шире диапазона промежуточных, у математики наоборот\n",
        "- Диапазон оценок по математике шире, чем по португальскому"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 500
        },
        "id": "g0FRI3VpwK8H",
        "outputId": "143d8fbf-b02a-4fb9-f6d5-4d0a2f27ab55"
      },
      "source": [
        "plt.figure(figsize=(12, 8))\n",
        "sns.histplot(data=clean_data[clean_data['Subject'] == 'Math'][['G1', 'G2', 'G3']], kde=True, multiple='stack')"
      ],
      "execution_count": 751,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fb40de495d0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 751
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hO4nOlGiYr9Z"
      },
      "source": [
        "- Итоговые и промежуточные оценки очень сильно коррелируют между собой, как можно видеть на графике выше\n",
        "- Разница чуть более заметна для оценок в диапазоне примерно от 5 до 8; здесь доминируют оценки за 1 семестр, т.е. в первый семестр в этом диапазоне их было больше, чем во втором семестр и в итоге\n",
        "- Относительно предыдущего вывода, вторым по численности в этом диапазоне был 2 семестр, а меньше всего оценок в этом диапазоне в итоговых оценках"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 500
        },
        "id": "eItZrVr4y7cm",
        "outputId": "32654a59-4518-4c82-846e-33a2f6ccd690"
      },
      "source": [
        "plt.figure(figsize=(12, 8))\n",
        "sns.histplot(data=clean_data[clean_data['Subject'] == 'Por'][['G1', 'G2', 'G3']], kde=True, multiple='stack')"
      ],
      "execution_count": 752,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fb40cff5850>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 752
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 864x576 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "NuWjcX-CZzx7"
      },
      "source": [
        "- Здесь статистика схожа с графиком распределения для математики, но основные расхождения оценок в промежуточных и итоговых оценках лежат в диапазоне от 5 до 10, что чуть больше, чем для португальского\n",
        "- В диапазоне от 14 до 16 итоговый балл чуть выше, чем в промежуточном оценивании"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7njZi_8b7O_E"
      },
      "source": [
        "\r\n",
        "\r\n",
        "- В общем, по графикам плотности распределения можно сказать, что в среднем наиболее высокими оказались итоговые оценки по математике и оценки по результатам второго полугодия по португальскому.\r\n",
        "\r\n"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7hr0jNRAqhlk"
      },
      "source": [
        "### Задание 6 \n",
        "  \n",
        "Оцените попарные зависимости числовых признаков.  \n",
        "Рассчитайте и провизуализируйте матрицу корреляций. Дайте содержательный комментарий."
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 503
        },
        "id": "FrROFs0wqhll",
        "outputId": "84b675b5-902a-459f-dc22-abefc89e2bae"
      },
      "source": [
        "plt.figure(figsize=(10, 8))\n",
        "sns.heatmap(clean_data.select_dtypes('int64').corr())"
      ],
      "execution_count": 753,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fb40cf8ce10>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 753
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x576 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "J3loSjySKXQ0"
      },
      "source": [
        "- Очевидно, что между признаками промежуточных и итоговых оценок высокая корреляция\n",
        "- Удивительно, но достаточно сильно коррелирует возраст с количеством завалов. Возможно, люди с возрастом начинают более серьёзно подходить к делу, возможно и наоборот. \n",
        "- А может они просто учатся дольше и успели накопить больше долгов, это уже надо отдельно смотреть :)\n",
        "- Негативная корреляция (обратная зависимость) существует и между оценками и количеством завалов, что достаточно логично"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "4ZxBTal2qhll"
      },
      "source": [
        "### Задание 7  \n",
        "  \n",
        "Оцените попарные зависимости нечисловых (категориальных) признаков, дайте содержательный комментарий. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 762
        },
        "id": "jj7UCxp_qhll",
        "outputId": "f4b4db8c-3f33-47cc-e187-ae92827b1f3e"
      },
      "source": [
        "plt.figure(figsize=(16, 12))\n",
        "catvars = enc_data.select_dtypes('category').astype(int)\n",
        "sns.heatmap(catvars.corr())"
      ],
      "execution_count": 754,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fb40cdd8490>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 754
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1152x864 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "7a8IOtqEKYWv"
      },
      "source": [
        "- Сильно коррелируют предмет и наличие платных дополнительных занятий\n",
        "- Также наблюдается сильная корреляция между местом проживания ученика и выбранной школой, ибо люди выбирают, где им удобнее учиться\n",
        "- Время на дорогу зависит от школы и места проживания, что связано с прошлым пунктом\n",
        "- Количество выпиваемого алкоголя и состояние здоровья связано с полом респондента!\n",
        "- Количество времени, проведённого с друзьями, зависит от количества свободного времени после занятий\n",
        "- Количество выпиваемого алкоголя зависит от количества времени, проведённого с друзьями\n",
        "- Работа матери зависит от её образования\n",
        "- Наблюдается отрицательная корреляция между образованием родителей и выбранной школой\n",
        "- Отрицательная корреляция наблюдается и между выбранной школой и предметом\n",
        "- Наличие интернета и планы получать высшее образование также подвержены отрицательной корреляции с образованием родителей"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "dOCyCk6tqhll"
      },
      "source": [
        "### Задание  8  \n",
        "  \n",
        "* Оцените, есть ли разница в употреблении алкоголя между парнями и девушками. Кто пьет больше?\n",
        "* Оцените связь употребления алкоголя с признаками: address, Pstatus, higher, romantic, freetime. \n",
        "* Оцените связь употребления алкоголя с уровнем образования опекуна (для случаев, где это известно).\n",
        "* Сравните употребление алкоголя для 50 лучших учеников и 50 худших (по `G3`). Выполните сравнение отдельно для каждого предмета (`Math` и `Por`). Дайте содержательный комментарий.  \n",
        "  \n",
        "**Tip:** для выполнения задания пригодятся боксплоты и гистограммы. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 292
        },
        "id": "ZrpWRpG6qhlm",
        "outputId": "265e9023-f47d-4259-dff9-d1a8680c1de1"
      },
      "source": [
        "# Разница в употреблении алкоголя между парнями и девушками\n",
        "catvars[['sex', 'Dalc', 'Walc']].groupby('sex').mean().plot(kind='bar')"
      ],
      "execution_count": 755,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fb40dc5d9d0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 755
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "UUjNjGFO35CX"
      },
      "source": [
        "- Здесь **0** -- **F**, а **1** -- **M**\n",
        "- Тогда отсюда следует, что мужчины в среднем пьют значительно больше как в будние, так и в выходные дни"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "-TcfZ4Ua4Tth"
      },
      "source": [
        "data_8 = enc_data[['address', 'Pstatus', 'higher', 'romantic', 'freetime']].astype(int)"
      ],
      "execution_count": 756,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Npq4uZ-O7Zpt"
      },
      "source": [
        "alc_features = (data_8.corrwith(enc_data['Dalc']) + data_8.corrwith(enc_data['Walc'])).sort_values()"
      ],
      "execution_count": 757,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 316
        },
        "id": "m4Z8A9Rf8MEm",
        "outputId": "df7e8ecd-1a30-4b72-efa8-d89ebde5639d"
      },
      "source": [
        "# Признаки, коррелирующие с употреблением алкоголя отсортированы в порядке значимости\n",
        "# Самый коррелирующий признак -- количество свободного времени\n",
        "# Наименее коррелирующий -- состояние в романтических отношениях\n",
        "alc_features.plot(kind='bar')"
      ],
      "execution_count": 758,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fb40cc9a4d0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 758
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 243
        },
        "id": "GCRxmKG5eKcr",
        "outputId": "251745a4-1a20-4f39-d8c5-a4d287214677"
      },
      "source": [
        "# Худшие студенты пьют достаточно много алкоголя\n",
        "plt.figure(figsize=(5,3))\n",
        "enc_data[['Dalc', 'Walc']] = enc_data[['Dalc', 'Walc']].astype(int)\n",
        "worst_students = enc_data.nsmallest(50, 'G3')\n",
        "sns.boxplot(data=worst_students[['Dalc','Walc']], orient='h').set_title('Употребление алкоголя 50 худшими студентами')"
      ],
      "execution_count": 759,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'Употребление алкоголя 50 худшими студентами')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 759
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 360x216 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 243
        },
        "id": "Zr2d24bieXK0",
        "outputId": "128534d4-addd-4f9f-b27d-b2d24a0b2740"
      },
      "source": [
        "# Лучше студенты почти не пьют алкоголь\n",
        "plt.figure(figsize=(5,3))\n",
        "best_students = data.nlargest(50, 'G3')\n",
        "sns.boxplot(data=best_students[['Dalc','Walc']], orient='h').set_title('Употребление алкоголя 50 лучшими студентами')"
      ],
      "execution_count": 760,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Text(0.5, 1.0, 'Употребление алкоголя 50 лучшими студентами')"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 760
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 360x216 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1Y7cgj6fqhlm"
      },
      "source": [
        "### Задание 9     \n",
        "  \n",
        "Оцените зависимость итоговой оценки от следующих признаков: \n",
        "  \n",
        "* Пола\n",
        "* Времени в пути от дома до школы\n",
        "* Наличия дополнительной помощи в учебе (от членов семьи или вообще)\n",
        "* Посещения платных дополнительных занятий\n",
        "* Наличия интернета\n",
        "* Наличия романтических отношений\n",
        "* Употребления алкоголя  "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        },
        "id": "h4D7cO4nqhlm",
        "outputId": "7196eea9-1d42-4984-ac99-01d9fae4136f"
      },
      "source": [
        "enc_data[['sex', 'traveltime', 'famsup', 'paid', 'internet', 'romantic']] = enc_data[\n",
        "  ['sex', 'traveltime', 'famsup', 'paid', 'internet', 'romantic']\n",
        "  ].astype(int)\n",
        "\n",
        "corr_9 = enc_data[\n",
        "                  ['sex', 'traveltime', 'famsup', 'paid', 'internet', 'romantic', 'Dalc', 'Walc']\n",
        "                  ].corrwith(enc_data['G3']).sort_values()\n",
        "\n",
        "corr_9.plot(kind='barh')"
      ],
      "execution_count": 761,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fb40cae4bd0>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 761
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "YaT9APRA1PfV"
      },
      "source": [
        "- Итоговая оценка находится в наименьшей обратной зависимости от пола и обратной зависимости от остальных признаков, из которых наиболее существенным является употребление алкоголя."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "i6-TWltHqhlm"
      },
      "source": [
        "### Задание 10  \n",
        "  \n",
        "Добавьте в данные новый признак - итоговая оценка по 4-х балльной шкале:  \n",
        "\n",
        "* Отлично: 18 <= `G3` <= 20\n",
        "* Хорошо: 14 <= `G3` <= 17\n",
        "* Удовлетворительно: 8 <= `G3` <= 13\n",
        "* Неудовлетворительно: `G3` < 8  \n",
        "  \n",
        "Оцените распределение нового признака в зависимости от признака `reason` (причина выбора школы). Дайте содержательный комментарий. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0r57rg4zqhln"
      },
      "source": [
        "def get_gpa(score):\n",
        "  score = round(score)\n",
        "  \n",
        "  if 18 <= score <= 20:\n",
        "    return 'Отлично'\n",
        "  if 14 <= score <= 17:\n",
        "    return 'Хорошо'\n",
        "  if 8 <= score <= 13:\n",
        "    return 'Удовлетворительно'\n",
        "  if score < 8:\n",
        "    return 'Неудовлетворительно'"
      ],
      "execution_count": 762,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "FDG9SeHEX_et"
      },
      "source": [
        "clean_data['GPA'] = clean_data['G3'].apply(get_gpa)\n",
        "enc_data['GPA'], _ = pd.factorize(clean_data['GPA'])"
      ],
      "execution_count": 763,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 217
        },
        "id": "GFkkCTBPuWgM",
        "outputId": "7e5bbc97-5c49-4c13-e745-6dd221b990fd"
      },
      "source": [
        "clean_data[clean_data['GPA'] == 'Удовлетворительно'].head()"
      ],
      "execution_count": 764,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>ID</th>\n",
              "      <th>Subject</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>reason</th>\n",
              "      <th>guardian</th>\n",
              "      <th>traveltime</th>\n",
              "      <th>studytime</th>\n",
              "      <th>failures</th>\n",
              "      <th>schoolsup</th>\n",
              "      <th>famsup</th>\n",
              "      <th>paid</th>\n",
              "      <th>activities</th>\n",
              "      <th>nursery</th>\n",
              "      <th>higher</th>\n",
              "      <th>internet</th>\n",
              "      <th>romantic</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>cheating</th>\n",
              "      <th>G1</th>\n",
              "      <th>G2</th>\n",
              "      <th>G3</th>\n",
              "      <th>GPA</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>102965</td>\n",
              "      <td>Por</td>\n",
              "      <td>GP</td>\n",
              "      <td>M</td>\n",
              "      <td>16</td>\n",
              "      <td>U</td>\n",
              "      <td>LE3</td>\n",
              "      <td>T</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>health</td>\n",
              "      <td>services</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>3</td>\n",
              "      <td>5</td>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>yes</td>\n",
              "      <td>9</td>\n",
              "      <td>8</td>\n",
              "      <td>10</td>\n",
              "      <td>Удовлетворительно</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>103131</td>\n",
              "      <td>Por</td>\n",
              "      <td>GP</td>\n",
              "      <td>F</td>\n",
              "      <td>16</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>home</td>\n",
              "      <td>mother</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>2</td>\n",
              "      <td>yes</td>\n",
              "      <td>12</td>\n",
              "      <td>13</td>\n",
              "      <td>13</td>\n",
              "      <td>Удовлетворительно</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>103144</td>\n",
              "      <td>Por</td>\n",
              "      <td>GP</td>\n",
              "      <td>M</td>\n",
              "      <td>18</td>\n",
              "      <td>U</td>\n",
              "      <td>LE3</td>\n",
              "      <td>T</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>services</td>\n",
              "      <td>services</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>no</td>\n",
              "      <td>11</td>\n",
              "      <td>11</td>\n",
              "      <td>12</td>\n",
              "      <td>Удовлетворительно</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>105257</td>\n",
              "      <td>Por</td>\n",
              "      <td>MS</td>\n",
              "      <td>F</td>\n",
              "      <td>18</td>\n",
              "      <td>U</td>\n",
              "      <td>GT3</td>\n",
              "      <td>T</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>other</td>\n",
              "      <td>other</td>\n",
              "      <td>course</td>\n",
              "      <td>mother</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>6</td>\n",
              "      <td>no</td>\n",
              "      <td>11</td>\n",
              "      <td>12</td>\n",
              "      <td>9</td>\n",
              "      <td>Удовлетворительно</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>108089</td>\n",
              "      <td>Por</td>\n",
              "      <td>MS</td>\n",
              "      <td>M</td>\n",
              "      <td>16</td>\n",
              "      <td>R</td>\n",
              "      <td>GT3</td>\n",
              "      <td>T</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>other</td>\n",
              "      <td>health</td>\n",
              "      <td>other</td>\n",
              "      <td>mother</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>yes</td>\n",
              "      <td>no</td>\n",
              "      <td>no</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>5</td>\n",
              "      <td>4</td>\n",
              "      <td>no</td>\n",
              "      <td>9</td>\n",
              "      <td>10</td>\n",
              "      <td>11</td>\n",
              "      <td>Удовлетворительно</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       ID Subject school sex  age address famsize Pstatus Medu Fedu      Mjob  \\\n",
              "2  102965     Por     GP   M   16       U     LE3       T    1    2    health   \n",
              "4  103131     Por     GP   F   16       U     GT3       T    1    1   at_home   \n",
              "5  103144     Por     GP   M   18       U     LE3       T    3    1  services   \n",
              "6  105257     Por     MS   F   18       U     GT3       T    1    1     other   \n",
              "9  108089     Por     MS   M   16       R     GT3       T    3    4     other   \n",
              "\n",
              "       Fjob  reason guardian traveltime  studytime  failures schoolsup famsup  \\\n",
              "2  services  course   mother          2          1         2        no     no   \n",
              "4     other    home   mother          2          1         0        no    yes   \n",
              "5  services  course   mother          2          1         0        no     no   \n",
              "6     other  course   mother          2          2         0        no     no   \n",
              "9    health   other   mother          3          2         0        no     no   \n",
              "\n",
              "  paid activities nursery higher internet romantic famrel freetime goout Dalc  \\\n",
              "2   no         no      no    yes      yes       no      4        4     5    3   \n",
              "4   no         no     yes    yes       no       no      4        3     2    1   \n",
              "5   no        yes     yes    yes      yes      yes      3        3     4    4   \n",
              "6   no        yes     yes    yes       no       no      1        1     1    1   \n",
              "9   no         no      no    yes       no       no      3        4     5    1   \n",
              "\n",
              "  Walc health  absences cheating  G1  G2  G3                GPA  \n",
              "2    5      5         0      yes   9   8  10  Удовлетворительно  \n",
              "4    4      5         2      yes  12  13  13  Удовлетворительно  \n",
              "5    5      4         2       no  11  11  12  Удовлетворительно  \n",
              "6    1      5         6       no  11  12   9  Удовлетворительно  \n",
              "9    2      5         4       no   9  10  11  Удовлетворительно  "
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 764
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 217
        },
        "id": "zj_QnqUXr0xh",
        "outputId": "0ee54bec-d20d-48c6-9ac8-ec0298f07635"
      },
      "source": [
        "enc_data[enc_data['GPA'] == 1].head()"
      ],
      "execution_count": 765,
      "outputs": [
        {
          "output_type": "execute_result",
          "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>ID</th>\n",
              "      <th>Subject</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>reason</th>\n",
              "      <th>guardian</th>\n",
              "      <th>traveltime</th>\n",
              "      <th>studytime</th>\n",
              "      <th>failures</th>\n",
              "      <th>schoolsup</th>\n",
              "      <th>famsup</th>\n",
              "      <th>paid</th>\n",
              "      <th>activities</th>\n",
              "      <th>nursery</th>\n",
              "      <th>higher</th>\n",
              "      <th>internet</th>\n",
              "      <th>romantic</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>cheating</th>\n",
              "      <th>G1</th>\n",
              "      <th>G2</th>\n",
              "      <th>G3</th>\n",
              "      <th>GPA</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>102965</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>16</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>3</td>\n",
              "      <td>5</td>\n",
              "      <td>5</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>9</td>\n",
              "      <td>8</td>\n",
              "      <td>10</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>103131</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>16</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>4</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>12</td>\n",
              "      <td>13</td>\n",
              "      <td>13</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>103144</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>18</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>4</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>11</td>\n",
              "      <td>11</td>\n",
              "      <td>12</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>105257</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>18</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>5</td>\n",
              "      <td>6</td>\n",
              "      <td>0</td>\n",
              "      <td>11</td>\n",
              "      <td>12</td>\n",
              "      <td>9</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>108089</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>1</td>\n",
              "      <td>16</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>3</td>\n",
              "      <td>1</td>\n",
              "      <td>3</td>\n",
              "      <td>2</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>1</td>\n",
              "      <td>0</td>\n",
              "      <td>3</td>\n",
              "      <td>4</td>\n",
              "      <td>5</td>\n",
              "      <td>1</td>\n",
              "      <td>2</td>\n",
              "      <td>5</td>\n",
              "      <td>4</td>\n",
              "      <td>0</td>\n",
              "      <td>9</td>\n",
              "      <td>10</td>\n",
              "      <td>11</td>\n",
              "      <td>1</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       ID Subject school  sex  age address famsize Pstatus Medu Fedu Mjob  \\\n",
              "2  102965       0      0    1   16       0       1       1    1    2    2   \n",
              "4  103131       0      0    0   16       0       0       1    1    1    3   \n",
              "5  103144       0      0    1   18       0       1       1    3    1    0   \n",
              "6  105257       0      1    0   18       0       0       1    1    1    1   \n",
              "9  108089       0      1    1   16       1       0       1    3    4    1   \n",
              "\n",
              "  Fjob reason guardian  traveltime studytime  failures schoolsup  famsup  \\\n",
              "2    1      0        1           2         1         2         0       0   \n",
              "4    0      1        1           2         1         0         0       1   \n",
              "5    1      0        1           2         1         0         0       0   \n",
              "6    0      0        1           2         2         0         0       0   \n",
              "9    2      3        1           3         2         0         0       0   \n",
              "\n",
              "   paid activities nursery higher  internet  romantic famrel freetime goout  \\\n",
              "2     0          0       1      0         0         0      4        4     5   \n",
              "4     0          0       0      0         1         0      4        3     2   \n",
              "5     0          1       0      0         0         1      3        3     4   \n",
              "6     0          1       0      0         1         0      1        1     1   \n",
              "9     0          0       1      0         1         0      3        4     5   \n",
              "\n",
              "   Dalc  Walc health  absences cheating  G1  G2  G3  GPA  \n",
              "2     3     5      5         0        1   9   8  10    1  \n",
              "4     1     4      5         2        1  12  13  13    1  \n",
              "5     4     5      4         2        0  11  11  12    1  \n",
              "6     1     1      5         6        0  11  12   9    1  \n",
              "9     1     2      5         4        0   9  10  11    1  "
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 765
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 281
        },
        "id": "sWxi3nH4ZoxL",
        "outputId": "9494801d-9e86-4e70-964f-fa40d1e1d164"
      },
      "source": [
        "enc_data['reason'] = enc_data['reason'].astype(int)\n",
        "enc_data[['GPA', 'reason']].plot.kde()"
      ],
      "execution_count": 766,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "<matplotlib.axes._subplots.AxesSubplot at 0x7fb40ca56390>"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 766
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 432x288 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HOV8UKSaKhPA"
      },
      "source": [
        "- Распределение этих двух признаков достаточно сильно совпадает в случаях, когда ученики имеют хорошую успеваемость и выбрали школу из-за читаемых курсов, а также в случае отличной успеваемости и иных причин выбора школы\n",
        "- Но достаточно сильное различие в распределении наблюдается в случае с выбором школы из-за близости к дому и удовлетворительной успеваемостью"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "rW89tuigqhln"
      },
      "source": [
        "### Задание 12  \n",
        "  \n",
        "Предложите простую модель для определения (предсказания) итоговой оценки студента по 4-х балльной системе. Оцените точность предложенной модели. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "bgeBE37uqhlo",
        "outputId": "0b4fc276-c8c7-45ad-ca4d-d273d5710a0a"
      },
      "source": [
        "from sklearn.linear_model import LogisticRegression\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.metrics import classification_report\n",
        "\n",
        "X_train, X_test, y_train, y_test = train_test_split(\n",
        "    enc_data.drop(['ID', 'G1', 'G2', 'G3'], axis=1), enc_data['GPA'])\n",
        "\n",
        "lr = LogisticRegression(max_iter=10000)\n",
        "lr.fit(X_train, y_train)"
      ],
      "execution_count": 767,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
              "                   intercept_scaling=1, l1_ratio=None, max_iter=10000,\n",
              "                   multi_class='auto', n_jobs=None, penalty='l2',\n",
              "                   random_state=None, solver='lbfgs', tol=0.0001, verbose=0,\n",
              "                   warm_start=False)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 767
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UAse4t52hjTC",
        "outputId": "d6529852-7de4-4599-b8f4-f19e983ead77"
      },
      "source": [
        "print(classification_report(lr.predict(X_test), y_test))"
      ],
      "execution_count": 768,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "              precision    recall  f1-score   support\n",
            "\n",
            "           0       1.00      1.00      1.00        57\n",
            "           1       1.00      0.95      0.97       162\n",
            "           2       0.38      0.56      0.45         9\n",
            "           3       0.60      1.00      0.75         6\n",
            "\n",
            "    accuracy                           0.95       234\n",
            "   macro avg       0.75      0.88      0.79       234\n",
            "weighted avg       0.97      0.95      0.96       234\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "1O138i5tyFPg"
      },
      "source": [
        "- Weighted F1-score составляет 96%, что достаточно неплохо, но было бы лучше попробовать бустинги типа xgboost или lightgbm, а не родимый логрег...\n",
        "- У меня возникли совсем большие проблемы со 2 классом. Вероятно, IQR определил слишком много выбросов и мы удалили много важного, надо было использовать z-score."
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "jhjhCtyRqhlo"
      },
      "source": [
        "### Задание * (бонусное)  \n",
        "  \n",
        "Дополнительное исследование данных и выводы положительно скажутся на оценке. "
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "uFaBoWCBqhlo"
      },
      "source": [
        "# Многие интересные выводы относительно этого написаны сразу под заданиями ;) \n",
        "# Не разрывать же изложение"
      ],
      "execution_count": null,
      "outputs": []
    }
  ]
}