{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "d3bd4846-56c3-4c76-97bf-b7ef53db174a",
   "metadata": {},
   "outputs": [],
   "source": [
    "## Выбрали из раздела Feature Selection SelectFromModel + LassoCV"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "90df1ff2-433e-473e-939d-4eacc10548bf",
   "metadata": {},
   "outputs": [],
   "source": [
    "#импортируем библиотеки\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "from sklearn.datasets import fetch_california_housing\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.linear_model import LassoCV, LinearRegression\n",
    "from sklearn.feature_selection import SelectFromModel\n",
    "from sklearn.metrics import mean_squared_error, r2_score\n",
    "\n",
    "%matplotlib inline\n",
    "plt.rcParams['figure.figsize'] = (12, 8)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "cd87e22c-0e8e-4e81-8319-484f391e23cc",
   "metadata": {},
   "source": [
    "## Цель задачи\n",
    "\n",
    "**Цель:** Продемонстрировать применение методов отбора признаков (SelectFromModel + LassoCV) на датасете California Housing для выявления ключевых факторов, влияющих на стоимость недвижимости.\n",
    "\n",
    "**Задачи:**\n",
    "- Обучить LassoCV для автоматического определения важности 8 признаков\n",
    "- Отобрать наиболее значимые признаки с помощью SelectFromModel\n",
    "- Сравнить качество линейной регрессии до и после отбора признаков\n",
    "- Визуализировать и интерпретировать результаты"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "63897308-6511-401a-a58a-7d1f59df62ea",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "==================================================\n",
      "ЧАСТЬ 1: California Housing\n",
      "==================================================\n",
      "Размер данных: (20640, 8)\n",
      "Признаки: ['MedInc', 'HouseAge', 'AveRooms', 'AveBedrms', 'Population', 'AveOccup', 'Latitude', 'Longitude']\n",
      "Целевая переменная (стоимость дома в $100,000):\n",
      "  Диапазон: [0.15, 5.00]\n"
     ]
    }
   ],
   "source": [
    "# California Housing Dataset\n",
    "\n",
    "# 1. ЗАГРУЗКА ДАННЫХ\n",
    "print(\"=\"*50)\n",
    "print(\"ЧАСТЬ 1: California Housing\")\n",
    "print(\"=\"*50)\n",
    "\n",
    "housing = fetch_california_housing()\n",
    "X1 = housing.data\n",
    "y1 = housing.target\n",
    "feature_names1 = housing.feature_names\n",
    "\n",
    "print(f\"Размер данных: {X1.shape}\")\n",
    "print(f\"Признаки: {feature_names1}\")\n",
    "print(f\"Целевая переменная (стоимость дома в $100,000):\")\n",
    "print(f\"  Диапазон: [{y1.min():.2f}, {y1.max():.2f}]\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "464680dd-3f16-448d-b6c8-2dd15b4f12f1",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Обучающая выборка: (16512, 8)\n",
      "Тестовая выборка: (4128, 8)\n"
     ]
    }
   ],
   "source": [
    "# Препроцессинг\n",
    "scaler1 = StandardScaler()\n",
    "X1_scaled = scaler1.fit_transform(X1)\n",
    "\n",
    "X1_train, X1_test, y1_train, y1_test = train_test_split(\n",
    "    X1_scaled, y1, test_size=0.2, random_state=42\n",
    ")\n",
    "\n",
    "print(f\"Обучающая выборка: {X1_train.shape}\")\n",
    "print(f\"Тестовая выборка: {X1_test.shape}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "a834e647-236f-4ab6-8975-6c81ee33ad74",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Оптимальный alpha: 0.000800\n",
      "\n",
      "Коэффициенты LassoCV:\n",
      "  MedInc: 0.8479\n",
      "  HouseAge: 0.1230\n",
      "  AveRooms: -0.2935\n",
      "  AveBedrms: 0.3588\n",
      "  Population: -0.0013\n",
      "  AveOccup: -0.0360\n",
      "  Latitude: -0.8879\n",
      "  Longitude: -0.8597\n"
     ]
    }
   ],
   "source": [
    "# LassoCV\n",
    "lasso1 = LassoCV(cv=5, random_state=42, max_iter=10000)\n",
    "lasso1.fit(X1_train, y1_train)\n",
    "\n",
    "print(f\"Оптимальный alpha: {lasso1.alpha_:.6f}\")\n",
    "print(\"\\nКоэффициенты LassoCV:\")\n",
    "for name, coef in zip(feature_names1, lasso1.coef_):\n",
    "    print(f\"  {name}: {coef:.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "937e88ef-508a-4e8c-8bee-2bebf759b175",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Выбрано признаков: 4 из 8\n",
      "Выбранные признаки: ['MedInc', 'AveBedrms', 'Latitude', 'Longitude']\n"
     ]
    }
   ],
   "source": [
    "# отбор через SelectFromModel\n",
    "sfm1 = SelectFromModel(lasso1, threshold='median')\n",
    "sfm1.fit(X1_train, y1_train)\n",
    "\n",
    "selected1 = sfm1.get_support()\n",
    "selected_names1 = [feature_names1[i] for i in range(len(feature_names1)) if selected1[i]]\n",
    "\n",
    "print(f\"\\nВыбрано признаков: {sum(selected1)} из {len(feature_names1)}\")\n",
    "print(f\"Выбранные признаки: {selected_names1}\")\n",
    "\n",
    "X1_train_selected = sfm1.transform(X1_train)\n",
    "X1_test_selected = sfm1.transform(X1_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "11d186dd-5f48-4dca-a121-720fc69e1ea4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "=== результаты California Housing ===\n",
      "Linear Regression (все признаки): R² = 0.5758\n",
      "Linear Regression (4 отобранных): R² = 0.5706\n",
      "LassoCV (все признаки): R² = 0.5767\n"
     ]
    }
   ],
   "source": [
    "# сравнение моделей\n",
    "# Модель на всех признаках\n",
    "lr1_full = LinearRegression()\n",
    "lr1_full.fit(X1_train, y1_train)\n",
    "y1_pred_full = lr1_full.predict(X1_test)\n",
    "\n",
    "# Модель только на отобранных признаках\n",
    "lr1_selected = LinearRegression()\n",
    "lr1_selected.fit(X1_train_selected, y1_train)\n",
    "y1_pred_selected = lr1_selected.predict(X1_test_selected)\n",
    "\n",
    "print(\"\\n=== результаты California Housing ===\")\n",
    "print(f\"Linear Regression (все признаки): R² = {r2_score(y1_test, y1_pred_full):.4f}\")\n",
    "print(f\"Linear Regression ({len(selected_names1)} отобранных): R² = {r2_score(y1_test, y1_pred_selected):.4f}\")\n",
    "print(f\"LassoCV (все признаки): R² = {r2_score(y1_test, lasso1.predict(X1_test)):.4f}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "a10ae3d1-1f13-4d2d-9d30-ceec249554f4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# визуализация\n",
    "plt.figure(figsize=(10, 6))\n",
    "coef_df = pd.DataFrame({'feature': feature_names1, 'coefficient': lasso1.coef_})\n",
    "coef_df = coef_df.sort_values('coefficient', key=abs, ascending=False)\n",
    "colors = ['red' if c < 0 else 'green' for c in coef_df['coefficient']]\n",
    "plt.barh(coef_df['feature'], coef_df['coefficient'], color=colors)\n",
    "plt.xlabel('Коэффициент')\n",
    "plt.title('Часть 1: Важность признаков (LassoCV на California Housing)')\n",
    "plt.axvline(x=0, color='black', linestyle='-', linewidth=0.5)\n",
    "plt.grid(True, alpha=0.3)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "58db0a76-8733-457d-a18c-60ee9428a1b4",
   "metadata": {},
   "source": [
    "## Интерпретация результатов (Feature Selection)\n",
    "\n",
    "- **Визуальный анализ коэффициентов**\n",
    "  - LassoCV обнулил коэффициенты для Population, HouseAge, AveRooms\n",
    "  - MedInc (доход) и Latitude/Longitude (геолокация) оказались ключевыми факторами\n",
    "\n",
    "- **Почему LassoCV выбрал эти признаки?**\n",
    "  - L1-регуляризация автоматически обнуляет неважные признаки\n",
    "  - SelectFromModel с порогом `median` оставляет признаки с коэффициентами выше медианы\n",
    "\n",
    "- **Сравнение качества**\n",
    "  - Все признаки (8): R² = 0.5758\n",
    "  - Отобранные (4): R² = 0.5706\n",
    "  - Потеря качества минимальна (менее 1%)\n",
    "\n",
    "- **Как улучшить?**\n",
    "  - Подобрать порог отбора (mean, 1.5*median)\n",
    "  - Попробовать RFE или SelectKBest\n",
    "\n",
    "**Вывод:**  \n",
    "LassoCV + SelectFromModel эффективно сокращают количество признаков (с 8 до 4) без значительной потери качества, упрощая модель и делая её более интерпретируемой."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "88011a02-9b03-4e7b-9316-60c7d7d53e61",
   "metadata": {},
   "source": [
    "## Реальный датасет Boston-house-price-data -> https://www.kaggle.com/datasets/arunjathari/bostonhousepricedata?resource=download"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "f276476c-19c9-4e9c-bb97-8fbc7bebe318",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "============================================================\n",
      "ЧАСТЬ 2: Boston Housing (загрузка из CSV)\n",
      "============================================================\n",
      "Размер датасета: (506, 14)\n",
      "\n",
      "Первые 5 строк данных:\n",
      "      CRIM    ZN  INDUS  CHAS    NOX     RM   AGE     DIS  RAD    TAX  \\\n",
      "0  0.00632  18.0   2.31     0  0.538  6.575  65.2  4.0900    1  296.0   \n",
      "1  0.02731   0.0   7.07     0  0.469  6.421  78.9  4.9671    2  242.0   \n",
      "2  0.02729   0.0   7.07     0  0.469  7.185  61.1  4.9671    2  242.0   \n",
      "3  0.03237   0.0   2.18     0  0.458  6.998  45.8  6.0622    3  222.0   \n",
      "4  0.06905   0.0   2.18     0  0.458  7.147  54.2  6.0622    3  222.0   \n",
      "\n",
      "   PTRATIO       B  LSTAT  MEDV  \n",
      "0     15.3  396.90   4.98  24.0  \n",
      "1     17.8  396.90   9.14  21.6  \n",
      "2     17.8  392.83   4.03  34.7  \n",
      "3     18.7  394.63   2.94  33.4  \n",
      "4     18.7  396.90   5.33  36.2  \n"
     ]
    }
   ],
   "source": [
    "# Boston Housing (pandas.read_csv)\n",
    "print(\"\\n\" + \"=\"*60)\n",
    "print(\"ЧАСТЬ 2: Boston Housing (загрузка из CSV)\")\n",
    "print(\"=\"*60)\n",
    "\n",
    "# Загружаем файл\n",
    "df = pd.read_csv('Boston-house-price-data.csv')\n",
    "\n",
    "print(f\"Размер датасета: {df.shape}\")\n",
    "print(f\"\\nПервые 5 строк данных:\")\n",
    "print(df.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a41eb6a1-56a5-4ed3-889c-3b35d7cec73a",
   "metadata": {},
   "source": [
    "## Цель задачи\n",
    "\n",
    "**Цель:** Продемонстрировать применение методов отбора признаков (SelectFromModel + LassoCV) на собственном датасете Boston Housing, загруженном через pandas.read_csv, для выявления ключевых факторов, влияющих на стоимость недвижимости.\n",
    "\n",
    "**Задачи:**\n",
    "- Загрузить данные из CSV-файла с помощью pandas.read_csv\n",
    "- Выполнить предобработку (масштабирование, разделение на выборки)\n",
    "- Обучить LassoCV для автоматического определения важности 13 признаков\n",
    "- Отобрать наиболее значимые признаки с помощью SelectFromModel (порог `median`)\n",
    "- Сравнить качество линейной регрессии до и после отбора признаков\n",
    "- Визуализировать и интерпретировать результаты"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "19a51828-3201-4b67-9f61-90521b9d902c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Информация о колонках:\n",
      "<class 'pandas.DataFrame'>\n",
      "RangeIndex: 506 entries, 0 to 505\n",
      "Data columns (total 14 columns):\n",
      " #   Column   Non-Null Count  Dtype  \n",
      "---  ------   --------------  -----  \n",
      " 0   CRIM     506 non-null    float64\n",
      " 1   ZN       506 non-null    float64\n",
      " 2   INDUS    506 non-null    float64\n",
      " 3   CHAS     506 non-null    int64  \n",
      " 4   NOX      506 non-null    float64\n",
      " 5   RM       506 non-null    float64\n",
      " 6   AGE      506 non-null    float64\n",
      " 7   DIS      506 non-null    float64\n",
      " 8   RAD      506 non-null    int64  \n",
      " 9   TAX      506 non-null    float64\n",
      " 10  PTRATIO  506 non-null    float64\n",
      " 11  B        506 non-null    float64\n",
      " 12  LSTAT    506 non-null    float64\n",
      " 13  MEDV     506 non-null    float64\n",
      "dtypes: float64(12), int64(2)\n",
      "memory usage: 55.5 KB\n",
      "None\n"
     ]
    }
   ],
   "source": [
    "print(f\"\\nИнформация о колонках:\")\n",
    "print(df.info())\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "33e81aac-5f7d-4447-813c-a3ed8c3b47bc",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "Признаки (13):\n",
      "   1. CRIM\n",
      "   2. ZN\n",
      "   3. INDUS\n",
      "   4. CHAS\n",
      "   5. NOX\n",
      "   6. RM\n",
      "   7. AGE\n",
      "   8. DIS\n",
      "   9. RAD\n",
      "  10. TAX\n",
      "  11. PTRATIO\n",
      "  12. B\n",
      "  13. LSTAT\n",
      "\n",
      "Целевая переменная: MEDV (медианная цена дома в $1000)\n",
      "Диапазон цен: от $5K до $50K\n",
      "Средняя цена: $23K\n"
     ]
    }
   ],
   "source": [
    "# Отделяем признаки и целевую переменную\n",
    "target_col = 'MEDV'  # MEDV = медианная стоимость дома\n",
    "X2 = df.drop(target_col, axis=1)\n",
    "y2 = df[target_col]\n",
    "feature_names2 = X2.columns.tolist()\n",
    "\n",
    "print(f\"\\nПризнаки ({len(feature_names2)}):\")\n",
    "for i, name in enumerate(feature_names2, 1):\n",
    "    print(f\"  {i:2d}. {name}\")\n",
    "\n",
    "print(f\"\\nЦелевая переменная: {target_col} (медианная цена дома в $1000)\")\n",
    "print(f\"Диапазон цен: от ${y2.min():.0f}K до ${y2.max():.0f}K\")\n",
    "print(f\"Средняя цена: ${y2.mean():.0f}K\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "f2926d2e-d126-4296-8115-591ec65c6e40",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "----------------------------------------\n",
      "ПРЕПРОЦЕССИНГ\n",
      "----------------------------------------\n",
      "До масштабирования: среднее ≈ 3.61, std = 8.60\n",
      "После масштабирования: среднее ≈ -0.00, std = 1.00\n",
      "\n",
      "Обучающая выборка: (404, 13)\n",
      "Тестовая выборка: (102, 13)\n"
     ]
    }
   ],
   "source": [
    "# Препроцессинг\n",
    "\n",
    "print(\"\\n\" + \"-\"*40)\n",
    "print(\"ПРЕПРОЦЕССИНГ\")\n",
    "print(\"-\"*40)\n",
    "\n",
    "# Масштабирование\n",
    "scaler2 = StandardScaler()\n",
    "X2_scaled = scaler2.fit_transform(X2)\n",
    "\n",
    "print(f\"До масштабирования: среднее ≈ {X2.iloc[:,0].mean():.2f}, std = {X2.iloc[:,0].std():.2f}\")\n",
    "print(f\"После масштабирования: среднее ≈ {X2_scaled[:,0].mean():.2f}, std = {X2_scaled[:,0].std():.2f}\")\n",
    "\n",
    "# Разделение на обучающую и тестовую выборки\n",
    "X2_train, X2_test, y2_train, y2_test = train_test_split(\n",
    "    X2_scaled, y2, test_size=0.2, random_state=42\n",
    ")\n",
    "\n",
    "print(f\"\\nОбучающая выборка: {X2_train.shape}\")\n",
    "print(f\"Тестовая выборка: {X2_test.shape}\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "c02fda06-cdb0-4791-9ca1-dca378b073b4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "----------------------------------------\n",
      "ОБУЧЕНИЕ LassoCV\n",
      "----------------------------------------\n",
      "Оптимальный коэффициент регуляризации alpha: 0.006833\n",
      "\n",
      "Коэффициенты LassoCV:\n",
      "  CRIM      : -0.9534 (важный)\n",
      "  ZN        : +0.6670 (важный)\n",
      "  INDUS     : +0.2147 (важный)\n",
      "  CHAS      : +0.7076 (важный)\n",
      "  NOX       : -1.9497 (важный)\n",
      "  RM        : +3.1201 (важный)\n",
      "  AGE       : -0.1613 (важный)\n",
      "  DIS       : -2.9986 (важный)\n",
      "  RAD       : +2.1568 (важный)\n",
      "  TAX       : -1.6663 (важный)\n",
      "  PTRATIO   : -1.9681 (важный)\n",
      "  B         : +1.1184 (важный)\n",
      "  LSTAT     : -3.6262 (важный)\n"
     ]
    }
   ],
   "source": [
    "# LassoCV\n",
    "print(\"\\n\" + \"-\"*40)\n",
    "print(\"ОБУЧЕНИЕ LassoCV\")\n",
    "print(\"-\"*40)\n",
    "\n",
    "lasso2 = LassoCV(cv=5, random_state=42, max_iter=10000)\n",
    "lasso2.fit(X2_train, y2_train)\n",
    "\n",
    "print(f\"Оптимальный коэффициент регуляризации alpha: {lasso2.alpha_:.6f}\")\n",
    "print(\"\\nКоэффициенты LassoCV:\")\n",
    "for name, coef in zip(feature_names2, lasso2.coef_):\n",
    "    if abs(coef) > 0.01:\n",
    "        print(f\"  {name:10s}: {coef:+.4f} (важный)\")\n",
    "    else:\n",
    "        print(f\"  {name:10s}: {coef:+.4f} (обнулен)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "65f21b0e-283c-4e3f-a553-531a7b20b2b5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "----------------------------------------\n",
      "ОТБОР ПРИЗНАКОВ (SelectFromModel)\n",
      "----------------------------------------\n",
      "Выбрано признаков: 7 из 13\n",
      "Выбранные признаки: ['NOX', 'RM', 'DIS', 'RAD', 'TAX', 'PTRATIO', 'LSTAT']\n"
     ]
    }
   ],
   "source": [
    "# отбор через SelectFromModel\n",
    "\n",
    "print(\"\\n\" + \"-\"*40)\n",
    "print(\"ОТБОР ПРИЗНАКОВ (SelectFromModel)\")\n",
    "print(\"-\"*40)\n",
    "\n",
    "sfm2 = SelectFromModel(lasso2, threshold='median')\n",
    "sfm2.fit(X2_train, y2_train)\n",
    "\n",
    "selected2 = sfm2.get_support()\n",
    "selected_names2 = [feature_names2[i] for i in range(len(feature_names2)) if selected2[i]]\n",
    "\n",
    "print(f\"Выбрано признаков: {sum(selected2)} из {len(feature_names2)}\")\n",
    "print(f\"Выбранные признаки: {selected_names2}\")\n",
    "\n",
    "# оставляем только отобранные признаки\n",
    "X2_train_selected = sfm2.transform(X2_train)\n",
    "X2_test_selected = sfm2.transform(X2_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "7315e9ed-002e-4e9d-a5a2-a8c9cdfb76f4",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "----------------------------------------\n",
      "СРАВНЕНИЕ КАЧЕСТВА МОДЕЛЕЙ\n",
      "----------------------------------------\n",
      "\n",
      "=== результаты Boston Housing ===\n",
      "\n",
      "Linear Regression (все 13 признаков):\n",
      "R²   = 0.6688\n",
      "RMSE = 4.93K\n",
      "\n",
      "Linear Regression (7 отобранных признаков):\n",
      "R²   = 0.6727\n",
      "RMSE = 4.90K\n",
      "\n",
      "LassoCV (все признаки):\n",
      "R²   = 0.6684\n",
      "RMSE = 4.93K\n",
      "\n",
      "ЭФФЕКТИВНОСТЬ ОТБОРА:\n",
      "Сокращение признаков: 13 → 7\n",
      "Потеря R²: -0.0039 (-0.6%)\n"
     ]
    }
   ],
   "source": [
    "# сравнение (до и после отбора)\n",
    "print(\"\\n\" + \"-\"*40)\n",
    "print(\"СРАВНЕНИЕ КАЧЕСТВА МОДЕЛЕЙ\")\n",
    "print(\"-\"*40)\n",
    "\n",
    "# Linear Regression на всех признаках\n",
    "lr2_full = LinearRegression()\n",
    "lr2_full.fit(X2_train, y2_train)\n",
    "y2_pred_full = lr2_full.predict(X2_test)\n",
    "r2_full = r2_score(y2_test, y2_pred_full)\n",
    "rmse_full = np.sqrt(mean_squared_error(y2_test, y2_pred_full))\n",
    "\n",
    "# Linear Regression только на отобранных признаках\n",
    "lr2_selected = LinearRegression()\n",
    "lr2_selected.fit(X2_train_selected, y2_train)\n",
    "y2_pred_selected = lr2_selected.predict(X2_test_selected)\n",
    "r2_selected = r2_score(y2_test, y2_pred_selected)\n",
    "rmse_selected = np.sqrt(mean_squared_error(y2_test, y2_pred_selected))\n",
    "\n",
    "# LassoCV\n",
    "y2_pred_lasso = lasso2.predict(X2_test)\n",
    "r2_lasso = r2_score(y2_test, y2_pred_lasso)\n",
    "rmse_lasso = np.sqrt(mean_squared_error(y2_test, y2_pred_lasso))\n",
    "\n",
    "print(\"\\n=== результаты Boston Housing ===\")\n",
    "print(f\"\\nLinear Regression (все {len(feature_names2)} признаков):\")\n",
    "print(f\"R²   = {r2_full:.4f}\")\n",
    "print(f\"RMSE = {rmse_full:.2f}K\")\n",
    "\n",
    "print(f\"\\nLinear Regression ({len(selected_names2)} отобранных признаков):\")\n",
    "print(f\"R²   = {r2_selected:.4f}\")\n",
    "print(f\"RMSE = {rmse_selected:.2f}K\")\n",
    "\n",
    "print(f\"\\nLassoCV (все признаки):\")\n",
    "print(f\"R²   = {r2_lasso:.4f}\")\n",
    "print(f\"RMSE = {rmse_lasso:.2f}K\")\n",
    "\n",
    "# Оценка эффективности отбора\n",
    "print(f\"\\nЭФФЕКТИВНОСТЬ ОТБОРА:\")\n",
    "print(f\"Сокращение признаков: {len(feature_names2)} → {len(selected_names2)}\")\n",
    "print(f\"Потеря R²: {r2_full - r2_selected:.4f} ({(r2_full - r2_selected)/r2_full*100:.1f}%)\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "f03eb06c-ac94-4095-8f26-cf28c0072492",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "----------------------------------------\n",
      "ВИЗУАЛИЗАЦИЯ\n",
      "----------------------------------------\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1500x500 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# визуализация\n",
    "\n",
    "print(\"\\n\" + \"-\"*40)\n",
    "print(\"ВИЗУАЛИЗАЦИЯ\")\n",
    "print(\"-\"*40)\n",
    "\n",
    "fig, axes = plt.subplots(1, 3, figsize=(15, 5))\n",
    "\n",
    "# График 1: Коэффициенты LassoCV\n",
    "coef_df2 = pd.DataFrame({'feature': feature_names2, 'coefficient': lasso2.coef_})\n",
    "coef_df2 = coef_df2.sort_values('coefficient', key=abs, ascending=False)\n",
    "colors2 = ['red' if c < 0 else 'green' for c in coef_df2['coefficient']]\n",
    "axes[0].barh(coef_df2['feature'], coef_df2['coefficient'], color=colors2)\n",
    "axes[0].axvline(x=0, color='black', linestyle='--', linewidth=0.5)\n",
    "axes[0].set_xlabel('Коэффициент')\n",
    "axes[0].set_title('Boston Housing: Важность признаков (LassoCV)\\nЗеленый (+) = рост цены, Красный (-) = снижение')\n",
    "axes[0].grid(True, alpha=0.3)\n",
    "\n",
    "# График 2: Предсказания vs Истина (модель на отобранных признаках)\n",
    "axes[1].scatter(y2_test, y2_pred_selected, alpha=0.5, s=20)\n",
    "axes[1].plot([y2_test.min(), y2_test.max()], [y2_test.min(), y2_test.max()], 'r--', lw=2)\n",
    "axes[1].set_xlabel('Истинные значения ($1000)')\n",
    "axes[1].set_ylabel('Предсказанные значения ($1000)')\n",
    "axes[1].set_title(f'Boston Housing: Предсказания\\n{len(selected_names2)} признаков, R² = {r2_selected:.4f}')\n",
    "axes[1].grid(True, alpha=0.3)\n",
    "\n",
    "# График 3: Сравнение ошибок (RMSE)\n",
    "models = ['Linear (все)', f'Linear ({len(selected_names2)})', 'LassoCV']\n",
    "rmse_values = [rmse_full, rmse_selected, rmse_lasso]\n",
    "colors_bar = ['blue', 'green', 'orange']\n",
    "axes[2].bar(models, rmse_values, color=colors_bar)\n",
    "axes[2].set_ylabel('RMSE ($1000)')\n",
    "axes[2].set_title('Сравнение ошибок моделей\\n(меньше = лучше)')\n",
    "axes[2].grid(True, alpha=0.3)\n",
    "\n",
    "plt.tight_layout()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7cad38f3-db19-427c-abd7-1d35e5e86ead",
   "metadata": {},
   "source": [
    "## 5. Интерпретация результатов (Boston Housing)\n",
    "\n",
    "- **Наблюдаемые эффекты**:\n",
    "  - LassoCV успешно обнулил коэффициенты для 6 неважных признаков, оставив 7 наиболее значимых\n",
    "  - Положительные коэффициенты (RM, RAD, ZN, CHAS) → увеличивают цену дома\n",
    "  - Отрицательные коэффициенты (LSTAT, DIS, PTRATIO, NOX, TAX, CRIM, AGE) → уменьшают цену дома\n",
    "  - Некоторые признаки (INDUS, B) получили коэффициенты, близкие к нулю → исключены\n",
    "\n",
    "- **Почему LassoCV выбрал именно эти 7 признаков?**\n",
    "  - L1-регуляризация автоматически обнуляет коэффициенты неважных признаков\n",
    "  - Порог `median` оставляет только признаки с коэффициентами выше медианы\n",
    "  - RM и LSTAT получили наибольшие коэффициенты → ключевые факторы\n",
    "\n",
    "- **Практическая значимость**:\n",
    "  - Сокращение признаков с 13 до 7 (почти в 2 раза) - экономия на сборе данных\n",
    "  - Модель становится проще и понятнее для заказчика\n",
    "  - Ключевые факторы (RM, LSTAT, PTRATIO) можно использовать для быстрой оценки\n",
    "\n",
    "**Вывод:** LassoCV + SelectFromModel эффективно сокращают количество признаков без потери качества, упрощая модель и делая её более интерпретируемой для практического использования."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "335e4a3f-56da-41bc-94cc-69cbd50d49b3",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.13.11"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}