colab/Вариант 3/notebook.ipynb

{
  "cells": [
    {
      "cell_type": "markdown",
      "id": "b833b51c",
      "metadata": {},
      "source": [
        "# Успеваемость учащихся (Вариант 3)\n",
        "\n",
        "Набор данных — синтетическое представление успеваемости: привычки учёбы, режим сна, социально-экономическое положение, посещаемость. Целевая переменная — **Grades** (оценки)."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 42,
      "id": "50172c1c",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "\n",
            "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m25.3\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m26.0.1\u001b[0m\n",
            "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpip install --upgrade pip\u001b[0m\n",
            "Note: you may need to restart the kernel to use updated packages.\n"
          ]
        }
      ],
      "source": [
        "%pip -q install pandas numpy matplotlib seaborn scikit-learn"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 43,
      "id": "dc8e6237",
      "metadata": {},
      "outputs": [],
      "source": [
        "import numpy as np\n",
        "import pandas as pd\n",
        "import matplotlib.pyplot as plt\n",
        "import seaborn as sns\n",
        "\n",
        "from sklearn.model_selection import train_test_split\n",
        "from sklearn.preprocessing import StandardScaler\n",
        "from sklearn.pipeline import Pipeline\n",
        "from sklearn.base import clone\n",
        "from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score\n",
        "from sklearn.linear_model import LinearRegression\n",
        "from sklearn.ensemble import RandomForestRegressor, GradientBoostingRegressor\n",
        "\n",
        "pd.set_option(\"display.max_columns\", 50)\n",
        "sns.set_style(\"whitegrid\")"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 44,
      "id": "df1da3e6",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Socioeconomic Score</th>\n",
              "      <th>Study Hours</th>\n",
              "      <th>Sleep Hours</th>\n",
              "      <th>Attendance (%)</th>\n",
              "      <th>Grades</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.95822</td>\n",
              "      <td>3.4</td>\n",
              "      <td>8.2</td>\n",
              "      <td>53.0</td>\n",
              "      <td>47.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.85566</td>\n",
              "      <td>3.2</td>\n",
              "      <td>5.9</td>\n",
              "      <td>55.0</td>\n",
              "      <td>35.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.68025</td>\n",
              "      <td>3.2</td>\n",
              "      <td>9.3</td>\n",
              "      <td>41.0</td>\n",
              "      <td>32.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.25936</td>\n",
              "      <td>3.2</td>\n",
              "      <td>8.2</td>\n",
              "      <td>47.0</td>\n",
              "      <td>34.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.60447</td>\n",
              "      <td>3.8</td>\n",
              "      <td>10.0</td>\n",
              "      <td>75.0</td>\n",
              "      <td>33.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Socioeconomic Score  Study Hours  Sleep Hours  Attendance (%)  Grades\n",
              "0              0.95822          3.4          8.2            53.0    47.0\n",
              "1              0.85566          3.2          5.9            55.0    35.0\n",
              "2              0.68025          3.2          9.3            41.0    32.0\n",
              "3              0.25936          3.2          8.2            47.0    34.0\n",
              "4              0.60447          3.8         10.0            75.0    33.0"
            ]
          },
          "execution_count": 44,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Загрузка датасета\n",
        "df = pd.read_csv(\"data.csv\")\n",
        "df.head()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "029f3bcf",
      "metadata": {},
      "source": [
        "## Очистка и предобработка\n",
        "\n",
        "- Удаляем дубликаты.\n",
        "- Проверяем пропуски и типы данных."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 45,
      "id": "cbff76d0",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Пропуски:\n",
            "Socioeconomic Score    0\n",
            "Study Hours            0\n",
            "Sleep Hours            0\n",
            "Attendance (%)         0\n",
            "Grades                 0\n",
            "dtype: int64\n",
            "\n",
            "<class 'pandas.DataFrame'>\n",
            "RangeIndex: 1388 entries, 0 to 1387\n",
            "Data columns (total 5 columns):\n",
            " #   Column               Non-Null Count  Dtype  \n",
            "---  ------               --------------  -----  \n",
            " 0   Socioeconomic Score  1388 non-null   float64\n",
            " 1   Study Hours          1388 non-null   float64\n",
            " 2   Sleep Hours          1388 non-null   float64\n",
            " 3   Attendance (%)       1388 non-null   float64\n",
            " 4   Grades               1388 non-null   float64\n",
            "dtypes: float64(5)\n",
            "memory usage: 54.3 KB\n",
            "\n",
            "Доля пропусков по столбцам:\n"
          ]
        },
        {
          "data": {
            "text/plain": [
              "Socioeconomic Score    0.0\n",
              "Study Hours            0.0\n",
              "Sleep Hours            0.0\n",
              "Attendance (%)         0.0\n",
              "Grades                 0.0\n",
              "dtype: float64"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        },
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Размер после очистки: 1388\n"
          ]
        }
      ],
      "source": [
        "df = df.drop_duplicates()\n",
        "print(\"Пропуски:\")\n",
        "print(df.isna().sum())\n",
        "print()\n",
        "df.info()\n",
        "print()\n",
        "print(\"Доля пропусков по столбцам:\")\n",
        "display(df.isna().mean().sort_values(ascending=False))\n",
        "df = df.dropna()\n",
        "print(\"Размер после очистки:\", len(df))"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "abc3d8c9",
      "metadata": {},
      "source": [
        "## Задание 1. Анализ исходных данных. Постановка задачи.\n"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 46,
      "id": "959f6247",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 600x400 with 1 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# График 1: распределение целевой переменной (Grades)\n",
        "fig, ax = plt.subplots(figsize=(6, 4))\n",
        "ax.hist(df[\"Grades\"], bins=30, edgecolor=\"black\", alpha=0.7)\n",
        "ax.set_xlabel(\"Grades (оценки)\")\n",
        "ax.set_ylabel(\"Частота\")\n",
        "ax.set_title(\"Распределение оценок учащихся\")\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "67234983",
      "metadata": {},
      "source": [
        "**Описание графика 1:** по горизонтали отложены оценки (Grades), по вертикали — частота. Чаще всего встречаются оценки 30-40 баллов."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 47,
      "id": "51b06bbe",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1200x400 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# График 2: зависимость оценок от часов учёбы и посещаемости\n",
        "fig, axes = plt.subplots(1, 2, figsize=(12, 4))\n",
        "axes[0].scatter(df[\"Study Hours\"], df[\"Grades\"], alpha=0.4, s=15)\n",
        "axes[0].set_xlabel(\"Study Hours (часы учёбы)\")\n",
        "axes[0].set_ylabel(\"Grades\")\n",
        "axes[0].set_title(\"Зависимость оценок от часов учёбы\")\n",
        "axes[1].scatter(df[\"Attendance (%)\"], df[\"Grades\"], alpha=0.4, s=15)\n",
        "axes[1].set_xlabel(\"Attendance (%) (посещаемость)\")\n",
        "axes[1].set_ylabel(\"Grades\")\n",
        "axes[1].set_title(\"Зависимость оценок от посещаемости\")\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "e0757c93",
      "metadata": {},
      "source": [
        "**Описание графика 2:** видно, что до 4.5 часов учёбы (предположительно в неделю) оценки ограничены 35-40 баллами, однако после 5 часов учёбы видна линейная нижня граница, например уделяя не менее 6 часов, нельзя получить балл ниже 40, а для 8 часов - 52 баллов.\n",
        "Посещаемость на оценки влияет схожим образом: до 80% большинство получают оценки ниже 40, а после видно улучшение оценок."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 48,
      "id": "a609d10f",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 600x500 with 2 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# График 3: корреляционная матрица числовых признаков и целевой переменной\n",
        "num_df = df.select_dtypes(include=[np.number])\n",
        "corr = num_df.corr()\n",
        "fig, ax = plt.subplots(figsize=(6, 5))\n",
        "sns.heatmap(corr, annot=True, fmt=\".2f\", cmap=\"coolwarm\", center=0, ax=ax, square=True)\n",
        "ax.set_title(\"Корреляции признаков и Grades\")\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "36894607",
      "metadata": {},
      "source": [
        "**Описание графика 3:** тепловая карта корреляций. Видимо, что сильнее всего на оценки влияет количество часов, уделяемое учёбе. Посещаемость и социально-экономические факторы также влияют, но слабее."
      ]
    },
    {
      "cell_type": "markdown",
      "id": "0ba40602",
      "metadata": {},
      "source": [
        "### Выводы о качестве датасета\n",
        "\n",
        "- **Очистка и предобработка:** выполнены удаление дубликатов и проверка пропусков.\n",
        "- **Качество:** признаки числовые, целевая переменная количественная; зависимости от факторов (часы учёбы, посещаемость, социально-экономический балл) позволяют строить регрессионную модель, т.к. данные не являются случайными.\n",
        "- **Необходимость новых переменных:** по результатам корреляций и графиков можно при необходимости ввести составной признак (например, «учебная активность» = часы учёбы × посещаемость) — это проверяется в Задании 3."
      ]
    },
    {
      "cell_type": "markdown",
      "id": "2839982a",
      "metadata": {},
      "source": [
        "## Задание 2. Построить модель прогнозирования оценки обучающегося\n",
        "\n",
        "Целевая переменная — **Grades** (количественная). Используем несколько моделей регрессии для сравнения точности на обучающей и тестовой выборках."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 49,
      "id": "7a5c03de",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Обучающая выборка: 1110 тестовая: 278\n"
          ]
        }
      ],
      "source": [
        "# Разделение на признаки и целевую переменную, train/test\n",
        "target = \"Grades\"\n",
        "feature_cols = [c for c in df.columns if c != target]\n",
        "X = df[feature_cols]\n",
        "y = df[target]\n",
        "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
        "print(\"Обучающая выборка:\", X_train.shape[0], \"тестовая:\", X_test.shape[0])"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 50,
      "id": "463c4576",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>model</th>\n",
              "      <th>split</th>\n",
              "      <th>MAE</th>\n",
              "      <th>RMSE</th>\n",
              "      <th>R2</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>LinearRegression</td>\n",
              "      <td>train</td>\n",
              "      <td>3.492656</td>\n",
              "      <td>4.523889</td>\n",
              "      <td>0.779763</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>LinearRegression</td>\n",
              "      <td>test</td>\n",
              "      <td>3.386175</td>\n",
              "      <td>4.390050</td>\n",
              "      <td>0.744667</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>RandomForest</td>\n",
              "      <td>train</td>\n",
              "      <td>0.363068</td>\n",
              "      <td>0.517689</td>\n",
              "      <td>0.997116</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>RandomForest</td>\n",
              "      <td>test</td>\n",
              "      <td>0.882644</td>\n",
              "      <td>1.189566</td>\n",
              "      <td>0.981252</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>GradientBoosting</td>\n",
              "      <td>train</td>\n",
              "      <td>0.718789</td>\n",
              "      <td>0.933955</td>\n",
              "      <td>0.990613</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>GradientBoosting</td>\n",
              "      <td>test</td>\n",
              "      <td>0.943528</td>\n",
              "      <td>1.285282</td>\n",
              "      <td>0.978114</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "              model  split       MAE      RMSE        R2\n",
              "0  LinearRegression  train  3.492656  4.523889  0.779763\n",
              "1  LinearRegression   test  3.386175  4.390050  0.744667\n",
              "2      RandomForest  train  0.363068  0.517689  0.997116\n",
              "3      RandomForest   test  0.882644  1.189566  0.981252\n",
              "4  GradientBoosting  train  0.718789  0.933955  0.990613\n",
              "5  GradientBoosting   test  0.943528  1.285282  0.978114"
            ]
          },
          "execution_count": 50,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "models = {\n",
        "    \"LinearRegression\": LinearRegression(),\n",
        "    \"RandomForest\": RandomForestRegressor(n_estimators=200, random_state=42),\n",
        "    \"GradientBoosting\": GradientBoostingRegressor(random_state=42),\n",
        "}\n",
        "\n",
        "def evaluate_models(X_tr, y_tr, X_te, y_te, model_dict):\n",
        "    \"\"\"Обучает модели, возвращает таблицу метрик на train и test.\"\"\"\n",
        "    rows = []\n",
        "    for name, model in model_dict.items():\n",
        "        pipe = Pipeline([\n",
        "            (\"scaler\", StandardScaler()),\n",
        "            (\"model\", clone(model)),\n",
        "        ])\n",
        "        pipe.fit(X_tr, y_tr)\n",
        "        for split, (X_s, y_s) in [(\"train\", (X_tr, y_tr)), (\"test\", (X_te, y_te))]:\n",
        "            pred = pipe.predict(X_s)\n",
        "            rows.append({\n",
        "                \"model\": name,\n",
        "                \"split\": split,\n",
        "                \"MAE\": mean_absolute_error(y_s, pred),\n",
        "                \"RMSE\": np.sqrt(mean_squared_error(y_s, pred)),\n",
        "                \"R2\": r2_score(y_s, pred),\n",
        "            })\n",
        "    return pd.DataFrame(rows)\n",
        "\n",
        "results = evaluate_models(X_train, y_train, X_test, y_test, models)\n",
        "results"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 51,
      "id": "4f79bc21",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr th {\n",
              "        text-align: left;\n",
              "    }\n",
              "\n",
              "    .dataframe thead tr:last-of-type th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr>\n",
              "      <th></th>\n",
              "      <th colspan=\"2\" halign=\"left\">MAE</th>\n",
              "      <th colspan=\"2\" halign=\"left\">R2</th>\n",
              "      <th colspan=\"2\" halign=\"left\">RMSE</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>split</th>\n",
              "      <th>test</th>\n",
              "      <th>train</th>\n",
              "      <th>test</th>\n",
              "      <th>train</th>\n",
              "      <th>test</th>\n",
              "      <th>train</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>model</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>GradientBoosting</th>\n",
              "      <td>0.943528</td>\n",
              "      <td>0.718789</td>\n",
              "      <td>0.978114</td>\n",
              "      <td>0.990613</td>\n",
              "      <td>1.285282</td>\n",
              "      <td>0.933955</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>LinearRegression</th>\n",
              "      <td>3.386175</td>\n",
              "      <td>3.492656</td>\n",
              "      <td>0.744667</td>\n",
              "      <td>0.779763</td>\n",
              "      <td>4.390050</td>\n",
              "      <td>4.523889</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>RandomForest</th>\n",
              "      <td>0.882644</td>\n",
              "      <td>0.363068</td>\n",
              "      <td>0.981252</td>\n",
              "      <td>0.997116</td>\n",
              "      <td>1.189566</td>\n",
              "      <td>0.517689</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                       MAE                  R2                RMSE          \n",
              "split                 test     train      test     train      test     train\n",
              "model                                                                       \n",
              "GradientBoosting  0.943528  0.718789  0.978114  0.990613  1.285282  0.933955\n",
              "LinearRegression  3.386175  3.492656  0.744667  0.779763  4.390050  4.523889\n",
              "RandomForest      0.882644  0.363068  0.981252  0.997116  1.189566  0.517689"
            ]
          },
          "execution_count": 51,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Сводная таблица метрик (train / test)\n",
        "pivot_results = results.pivot_table(index=\"model\", columns=\"split\", values=[\"MAE\", \"RMSE\", \"R2\"])\n",
        "pivot_results"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 52,
      "id": "9737f501",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "image/png": "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",
            "text/plain": [
              "<Figure size 1400x400 with 3 Axes>"
            ]
          },
          "metadata": {},
          "output_type": "display_data"
        }
      ],
      "source": [
        "# Визуализация метрик на тестовой выборке\n",
        "test_results = results[results[\"split\"] == \"test\"]\n",
        "fig, axes = plt.subplots(1, 3, figsize=(14, 4))\n",
        "sns.barplot(data=test_results, x=\"model\", y=\"MAE\", ax=axes[0])\n",
        "sns.barplot(data=test_results, x=\"model\", y=\"RMSE\", ax=axes[1])\n",
        "sns.barplot(data=test_results, x=\"model\", y=\"R2\", ax=axes[2])\n",
        "axes[0].set_title(\"MAE (test)\")\n",
        "axes[1].set_title(\"RMSE (test)\")\n",
        "axes[2].set_title(\"R2 (test)\")\n",
        "plt.tight_layout()\n",
        "plt.show()"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "5e66fccb",
      "metadata": {},
      "source": [
        "### Обоснование выбора метрик\n",
        "\n",
        "- **MAE** (Mean Absolute Error) — средняя абсолютная ошибка в тех же единицах, что и оценки; легко интерпретировать, устойчива к выбросам.\n",
        "- **RMSE** (Root Mean Squared Error) — корень из средней квадратичной ошибки; сильнее штрафует большие отклонения, полезна для оценки риска крупных промахов.\n",
        "- **R2** (коэффициент детерминации) — доля дисперсии целевой переменной, объяснённая моделью; 0 — модель не лучше предсказания средним, 1 — идеальное совпадение."
      ]
    },
    {
      "cell_type": "markdown",
      "id": "a909c402",
      "metadata": {},
      "source": [
        "## Задание 3. Новый признак и анализ изменения точности\n",
        "\n",
        "Формируем новый признак вручную: **Study effort** (учебная активность) = часы учёбы × посещаемость (нормализованная 0–1). Он объединяет два важных фактора успеваемости. Далее сравниваем метрики моделей до и после добавления этого признака."
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 53,
      "id": "ae3ee431",
      "metadata": {},
      "outputs": [
        {
          "name": "stdout",
          "output_type": "stream",
          "text": [
            "Новые признаки: ['Socioeconomic Score', 'Study Hours', 'Sleep Hours', 'Attendance (%)', 'Study_effort']\n"
          ]
        },
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Socioeconomic Score</th>\n",
              "      <th>Study Hours</th>\n",
              "      <th>Sleep Hours</th>\n",
              "      <th>Attendance (%)</th>\n",
              "      <th>Study_effort</th>\n",
              "      <th>Grades</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>0.95822</td>\n",
              "      <td>3.4</td>\n",
              "      <td>8.2</td>\n",
              "      <td>53.0</td>\n",
              "      <td>1.802</td>\n",
              "      <td>47.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>0.85566</td>\n",
              "      <td>3.2</td>\n",
              "      <td>5.9</td>\n",
              "      <td>55.0</td>\n",
              "      <td>1.760</td>\n",
              "      <td>35.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>0.68025</td>\n",
              "      <td>3.2</td>\n",
              "      <td>9.3</td>\n",
              "      <td>41.0</td>\n",
              "      <td>1.312</td>\n",
              "      <td>32.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>0.25936</td>\n",
              "      <td>3.2</td>\n",
              "      <td>8.2</td>\n",
              "      <td>47.0</td>\n",
              "      <td>1.504</td>\n",
              "      <td>34.0</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>0.60447</td>\n",
              "      <td>3.8</td>\n",
              "      <td>10.0</td>\n",
              "      <td>75.0</td>\n",
              "      <td>2.850</td>\n",
              "      <td>33.0</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "   Socioeconomic Score  Study Hours  Sleep Hours  Attendance (%)  \\\n",
              "0              0.95822          3.4          8.2            53.0   \n",
              "1              0.85566          3.2          5.9            55.0   \n",
              "2              0.68025          3.2          9.3            41.0   \n",
              "3              0.25936          3.2          8.2            47.0   \n",
              "4              0.60447          3.8         10.0            75.0   \n",
              "\n",
              "   Study_effort  Grades  \n",
              "0         1.802    47.0  \n",
              "1         1.760    35.0  \n",
              "2         1.312    32.0  \n",
              "3         1.504    34.0  \n",
              "4         2.850    33.0  "
            ]
          },
          "execution_count": 53,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Новый признак: учебная активность (часы учёбы × доля посещаемости)\n",
        "df_ext = df.copy()\n",
        "df_ext[\"Study_effort\"] = df_ext[\"Study Hours\"] * (df_ext[\"Attendance (%)\"] / 100.0)\n",
        "feature_cols_ext = feature_cols + [\"Study_effort\"]\n",
        "X_ext = df_ext[feature_cols_ext]\n",
        "y_ext = df_ext[target]\n",
        "X_train_ext, X_test_ext, y_train_ext, y_test_ext = train_test_split(\n",
        "    X_ext, y_ext, test_size=0.2, random_state=42\n",
        ")\n",
        "print(\"Новые признаки:\", feature_cols_ext)\n",
        "df_ext[feature_cols_ext + [target]].head()"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 54,
      "id": "27f43283",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>model</th>\n",
              "      <th>split</th>\n",
              "      <th>MAE</th>\n",
              "      <th>RMSE</th>\n",
              "      <th>R2</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>LinearRegression</td>\n",
              "      <td>train</td>\n",
              "      <td>3.236795</td>\n",
              "      <td>4.305121</td>\n",
              "      <td>0.800548</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>LinearRegression</td>\n",
              "      <td>test</td>\n",
              "      <td>3.217861</td>\n",
              "      <td>4.224914</td>\n",
              "      <td>0.763515</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>RandomForest</td>\n",
              "      <td>train</td>\n",
              "      <td>0.369018</td>\n",
              "      <td>0.529524</td>\n",
              "      <td>0.996983</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>RandomForest</td>\n",
              "      <td>test</td>\n",
              "      <td>0.897176</td>\n",
              "      <td>1.218256</td>\n",
              "      <td>0.980337</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>GradientBoosting</td>\n",
              "      <td>train</td>\n",
              "      <td>0.716615</td>\n",
              "      <td>0.928923</td>\n",
              "      <td>0.990714</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>GradientBoosting</td>\n",
              "      <td>test</td>\n",
              "      <td>0.940395</td>\n",
              "      <td>1.281435</td>\n",
              "      <td>0.978245</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "              model  split       MAE      RMSE        R2\n",
              "0  LinearRegression  train  3.236795  4.305121  0.800548\n",
              "1  LinearRegression   test  3.217861  4.224914  0.763515\n",
              "2      RandomForest  train  0.369018  0.529524  0.996983\n",
              "3      RandomForest   test  0.897176  1.218256  0.980337\n",
              "4  GradientBoosting  train  0.716615  0.928923  0.990714\n",
              "5  GradientBoosting   test  0.940395  1.281435  0.978245"
            ]
          },
          "execution_count": 54,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Обучаем те же модели на данных с новым признаком\n",
        "results_with_feature = evaluate_models(\n",
        "    X_train_ext, y_train_ext, X_test_ext, y_test_ext, models\n",
        ")\n",
        "results_with_feature"
      ]
    },
    {
      "cell_type": "code",
      "execution_count": 55,
      "id": "fd94163a",
      "metadata": {},
      "outputs": [
        {
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>MAE_before</th>\n",
              "      <th>MAE_after</th>\n",
              "      <th>RMSE_before</th>\n",
              "      <th>RMSE_after</th>\n",
              "      <th>R2_before</th>\n",
              "      <th>R2_after</th>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>model</th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "      <th></th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>LinearRegression</th>\n",
              "      <td>3.386175</td>\n",
              "      <td>3.217861</td>\n",
              "      <td>4.390050</td>\n",
              "      <td>4.224914</td>\n",
              "      <td>0.744667</td>\n",
              "      <td>0.763515</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>RandomForest</th>\n",
              "      <td>0.882644</td>\n",
              "      <td>0.897176</td>\n",
              "      <td>1.189566</td>\n",
              "      <td>1.218256</td>\n",
              "      <td>0.981252</td>\n",
              "      <td>0.980337</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>GradientBoosting</th>\n",
              "      <td>0.943528</td>\n",
              "      <td>0.940395</td>\n",
              "      <td>1.285282</td>\n",
              "      <td>1.281435</td>\n",
              "      <td>0.978114</td>\n",
              "      <td>0.978245</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                  MAE_before  MAE_after  RMSE_before  RMSE_after  R2_before  \\\n",
              "model                                                                         \n",
              "LinearRegression    3.386175   3.217861     4.390050    4.224914   0.744667   \n",
              "RandomForest        0.882644   0.897176     1.189566    1.218256   0.981252   \n",
              "GradientBoosting    0.943528   0.940395     1.285282    1.281435   0.978114   \n",
              "\n",
              "                  R2_after  \n",
              "model                       \n",
              "LinearRegression  0.763515  \n",
              "RandomForest      0.980337  \n",
              "GradientBoosting  0.978245  "
            ]
          },
          "execution_count": 55,
          "metadata": {},
          "output_type": "execute_result"
        }
      ],
      "source": [
        "# Сравнение метрик на тесте: без нового признака vs с признаком Study_effort\n",
        "tb = results[results[\"split\"] == \"test\"].set_index(\"model\")\n",
        "ta = results_with_feature[results_with_feature[\"split\"] == \"test\"].set_index(\"model\")\n",
        "comparison = pd.DataFrame({\n",
        "    \"MAE_before\": tb[\"MAE\"],\n",
        "    \"MAE_after\": ta[\"MAE\"],\n",
        "    \"RMSE_before\": tb[\"RMSE\"],\n",
        "    \"RMSE_after\": ta[\"RMSE\"],\n",
        "    \"R2_before\": tb[\"R2\"],\n",
        "    \"R2_after\": ta[\"R2\"],\n",
        "})\n",
        "comparison"
      ]
    },
    {
      "cell_type": "markdown",
      "id": "897f43ad",
      "metadata": {},
      "source": [
        "### Анализ изменения точности\n",
        "\n",
        "Добавление нового признака на основе уже существующих не принесло никаких изменений. Это связана с тем, что модель и без умеет учитывать все признаки, а новый столбец, полученный умножением двух других, новой информации не несёт."
      ]
    }
  ],
  "metadata": {
    "kernelspec": {
      "display_name": ".venv (3.14.2)",
      "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": "3.14.2"
    }
  },
  "nbformat": 4,
  "nbformat_minor": 5
}