{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Начало лабораторной работы"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['age', 'sex', 'bmi', 'children', 'smoker', 'region', 'charges'], dtype='object')\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "df = pd.read_csv(\"..//static//csv//Medical_insurance.csv\")\n",
    "print(df.columns)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Бизнес-цели\n",
    "\n",
    "1. Оптимизация тарифов:\n",
    "\n",
    "Цель: Разработка более точных и справедливых тарифов на страховку, основанных на индивидуальных рисках клиентов.\n",
    "\n",
    "Показатели успеха:\n",
    "\n",
    "Снижение оттока клиентов (уменьшение количества отказов от страховки).\n",
    "\n",
    "Увеличение прибыли за счет более точного ценообразования.\n",
    "\n",
    "Повышение удовлетворенности клиентов (опросы, отзывы).\n",
    "\n",
    "2. Оценка рисков:\n",
    "\n",
    "Цель: Оценка рисков для новых видов страхования или географических регионов.\n",
    "\n",
    "Показатели успеха:\n",
    "\n",
    "Снижение убытков от страховых случаев.\n",
    "\n",
    "Увеличение прибыли за счет выхода на новые рынки.\n",
    "\n",
    "Сокращение сроков разработки новых страховых продуктов."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Технические цели для каждой бизнес-цели:\n",
    "\n",
    "1. Оптимизация тарифов:\n",
    "\n",
    "* **Техническая цель:**  Разработка и внедрение модели машинного обучения, которая будет прогнозировать стоимость страховки с высокой точностью на основе данных о клиентах (возраст, пол, ИМТ, количество детей, статус курения, регион проживания).\n",
    "* **Ключевые задачи:**\n",
    "    * Сбор и подготовка данных (очистка, нормализация, обработка пропущенных значений).\n",
    "    * Исследование данных и выявление важных признаков.\n",
    "    * Выбор и обучение модели машинного обучения (линейная регрессия, случайный лес, градиентный бустинг и т.д.).\n",
    "    * Оценка качества модели (метрики: RMSE, MAE, R²).\n",
    "    * Разработка API для интеграции модели в существующие системы компании.\n",
    "    * Тестирование и развертывание модели в продакшн.\n",
    "    * Мониторинг и поддержка модели (обновление данных, переобучение модели).\n",
    "\n",
    "2. Оценка рисков:\n",
    "\n",
    "* **Техническая цель:**  Разработка модели оценки рисков для новых видов страхования или географических регионов, которая позволит определить потенциальные убытки и скорректировать тарифы соответствующим образом.\n",
    "* **Ключевые задачи:**\n",
    "    * Сбор и подготовка данных о новых видах страхования или регионах (исторические данные, демографическая информация, данные о рисках).\n",
    "    * Исследование данных и выявление закономерностей, связанных с рисками.\n",
    "    * Выбор и обучение модели машинного обучения для оценки рисков (классификация, регрессия).\n",
    "    * Оценка качества модели (метрики: точность, полнота, F1-мера, AUC-ROC).\n",
    "    * Разработка отчетов и дашбордов для визуализации результатов оценки рисков.\n",
    "    * Интеграция модели в процесс принятия решений о тарифах и страховых продуктах.\n",
    "    * Тестирование и развертывание модели в продакшн.\n",
    "    * Мониторинг и поддержка модели (обновление данных, переобучение модели).\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Размер обучающей выборки: (1940, 6)\n",
      "Размер контрольной выборки: (416, 6)\n",
      "Размер тестовой выборки: (416, 6)\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "\n",
    "# Загрузка данных\n",
    "df = pd.read_csv(\"..//static//csv//Medical_insurance.csv\")\n",
    "\n",
    "# Перемешивание данных\n",
    "df = df.sample(frac=1, random_state=42).reset_index(drop=True)\n",
    "\n",
    "# Разделение на признаки и целевую переменную\n",
    "X = df.drop('charges', axis=1)\n",
    "y = df['charges']\n",
    "\n",
    "# Разбиение на обучающую, контрольную и тестовую выборки\n",
    "X_train, X_temp, y_train, y_temp = train_test_split(X, y, test_size=0.3, random_state=42)\n",
    "X_val, X_test, y_val, y_test = train_test_split(X_temp, y_temp, test_size=0.5, random_state=42)\n",
    "\n",
    "# Вывод размеров выборок\n",
    "print(\"Размер обучающей выборки:\", X_train.shape)\n",
    "print(\"Размер контрольной выборки:\", X_val.shape)\n",
    "print(\"Размер тестовой выборки:\", X_test.shape)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1800x600 with 3 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "# Загрузка данных\n",
    "df = pd.read_csv(\"..//static//csv//Medical_insurance.csv\")\n",
    "\n",
    "# Перемешивание данных\n",
    "df = df.sample(frac=1, random_state=42).reset_index(drop=True)\n",
    "\n",
    "# Разделение на признаки и целевую переменную\n",
    "X = df.drop('charges', axis=1)\n",
    "y = df['charges']\n",
    "\n",
    "# Разбиение на обучающую, контрольную и тестовую выборки\n",
    "X_train, X_temp, y_train, y_temp = train_test_split(X, y, test_size=0.3, random_state=42)\n",
    "X_val, X_test, y_val, y_test = train_test_split(X_temp, y_temp, test_size=0.5, random_state=42)\n",
    "\n",
    "# Функция для оценки сбалансированности выборок\n",
    "def evaluate_balance(y_train, y_val, y_test):\n",
    "    plt.figure(figsize=(18, 6))\n",
    "    \n",
    "    plt.subplot(1, 3, 1)\n",
    "    sns.histplot(y_train, kde=True)\n",
    "    plt.title('Распределение целевой переменной в обучающей выборке')\n",
    "    \n",
    "    plt.subplot(1, 3, 2)\n",
    "    sns.histplot(y_val, kde=True)\n",
    "    plt.title('Распределение целевой переменной в контрольной выборке')\n",
    "    \n",
    "    plt.subplot(1, 3, 3)\n",
    "    sns.histplot(y_test, kde=True)\n",
    "    plt.title('Распределение целевой переменной в тестовой выборке')\n",
    "    \n",
    "    plt.show()\n",
    "\n",
    "# Оценка сбалансированности выборок\n",
    "evaluate_balance(y_train, y_val, y_test)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Процесс конструирования признаков для обеих задач:**\n",
    "1. **Анализ и очистка данных:**\n",
    "   - Проверить наличие пропущенных значений и дубликатов и обработать их (заполнение средним значением, медианой или удаление строк).\n",
    "\n",
    "2. **Кодирование категориальных признаков:**\n",
    "   - Применить One-Hot Encoding для категориальных признаков (`sex`, `smoker`, `region`).\n",
    "\n",
    "3. **Создание новых признаков:**\n",
    "   - **Возрастные группы:** Разделить возраст на группы (например, молодые, средний возраст, пожилые).\n",
    "   - **Индекс массы тела (ИМТ) группы:** Разделить ИМТ на группы (например, недостаточный вес, нормальный вес, избыточный вес, ожирение).\n",
    "   - **Количество детей:** Создать бинарный признак, указывающий, есть ли у клиента дети.\n",
    "   - **Комбинированные признаки:** Создать новые признаки, комбинируя существующие (например, возраст и ИМТ, возраст и статус курения).\n",
    "\n",
    "4. **Нормализация и стандартизация:**\n",
    "   - Применить нормализацию или стандартизацию к числовым признакам (`age`, `bmi`, `children`), чтобы привести их к одному масштабу.\n",
    "\n",
    "5. **Анализ важности признаков:**\n",
    "   - Удалить малозначимые признаки, чтобы упростить модель и улучшить ее производительность."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 1. Оптимизация тарифов"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Обучающая выборка после конструирования признаков:\n",
      "           age       bmi  children  sex_female  sex_male  smoker_no  \\\n",
      "1684  1.781292  0.374453 -0.907604        True     False       True   \n",
      "862  -0.083478 -0.570585 -0.085975        True     False       True   \n",
      "1992 -1.087585 -0.495147 -0.907604        True     False       True   \n",
      "889   0.705463 -0.586335 -0.907604        True     False       True   \n",
      "1362  0.777185 -0.680839 -0.907604       False      True       True   \n",
      "\n",
      "      smoker_yes  region_northeast  region_northwest  region_southeast  \\\n",
      "1684       False             False              True             False   \n",
      "862        False              True             False             False   \n",
      "1992       False             False             False              True   \n",
      "889        False             False             False              True   \n",
      "1362       False             False             False             False   \n",
      "\n",
      "      region_southwest  age_bin   age_bmi  \n",
      "1684             False      4.0  1.734561  \n",
      "862              False      1.0 -0.339153  \n",
      "1992             False      0.0 -1.055293  \n",
      "889              False      2.0  0.231109  \n",
      "1362              True      2.0  0.228540  \n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import OneHotEncoder, StandardScaler\n",
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.pipeline import Pipeline\n",
    "\n",
    "# Загрузка данных\n",
    "df = pd.read_csv(\"..//static//csv//Medical_insurance.csv\")\n",
    "\n",
    "# Перемешивание данных\n",
    "df = df.sample(frac=1, random_state=42).reset_index(drop=True)\n",
    "\n",
    "# Разделение на признаки и целевую переменную\n",
    "X = df.drop('charges', axis=1)\n",
    "y = df['charges']\n",
    "\n",
    "# Разбиение на обучающую, контрольную и тестовую выборки\n",
    "X_train, X_temp, y_train, y_temp = train_test_split(X, y, test_size=0.3, random_state=42)\n",
    "X_val, X_test, y_val, y_test = train_test_split(X_temp, y_temp, test_size=0.5, random_state=42)\n",
    "\n",
    "# Определение категориальных и числовых признаков\n",
    "categorical_features = ['sex', 'smoker', 'region']\n",
    "numerical_features = ['age', 'bmi', 'children']\n",
    "\n",
    "# Унитарное кодирование категориальных признаков (one-hot encoding)\n",
    "train_data_encoded = pd.get_dummies(X_train, columns=categorical_features)\n",
    "val_data_encoded = pd.get_dummies(X_val, columns=categorical_features)\n",
    "test_data_encoded = pd.get_dummies(X_test, columns=categorical_features)\n",
    "\n",
    "# Дискретизация числовых признаков (пример для возраста)\n",
    "age_bins = [18, 30, 40, 50, 60, 100]\n",
    "train_data_encoded['age_bin'] = pd.cut(train_data_encoded['age'], bins=age_bins, labels=False)\n",
    "val_data_encoded['age_bin'] = pd.cut(val_data_encoded['age'], bins=age_bins, labels=False)\n",
    "test_data_encoded['age_bin'] = pd.cut(test_data_encoded['age'], bins=age_bins, labels=False)\n",
    "\n",
    "# «Ручной» синтез признаков (пример: комбинированный признак возраст и ИМТ)\n",
    "train_data_encoded['age_bmi'] = train_data_encoded['age'] * train_data_encoded['bmi']\n",
    "val_data_encoded['age_bmi'] = val_data_encoded['age'] * val_data_encoded['bmi']\n",
    "test_data_encoded['age_bmi'] = test_data_encoded['age'] * test_data_encoded['bmi']\n",
    "\n",
    "# Масштабирование числовых признаков\n",
    "numerical_features = ['age', 'bmi', 'children', 'age_bmi']\n",
    "scaler = StandardScaler()\n",
    "train_data_encoded[numerical_features] = scaler.fit_transform(train_data_encoded[numerical_features])\n",
    "val_data_encoded[numerical_features] = scaler.transform(val_data_encoded[numerical_features])\n",
    "test_data_encoded[numerical_features] = scaler.transform(test_data_encoded[numerical_features])\n",
    "\n",
    "# Вывод результатов\n",
    "print(\"Обучающая выборка после конструирования признаков:\")\n",
    "print(train_data_encoded.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## 2. Оценка рисков"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Обучающая выборка после конструирования признаков:\n",
      "           age       bmi  children  sex_female  sex_male  smoker_no  \\\n",
      "1684  1.781292  0.374453 -0.907604        True     False       True   \n",
      "862  -0.083478 -0.570585 -0.085975        True     False       True   \n",
      "1992 -1.087585 -0.495147 -0.907604        True     False       True   \n",
      "889   0.705463 -0.586335 -0.907604        True     False       True   \n",
      "1362  0.777185 -0.680839 -0.907604       False      True       True   \n",
      "\n",
      "      smoker_yes  region_northeast  region_northwest  region_southeast  \\\n",
      "1684       False             False              True             False   \n",
      "862        False              True             False             False   \n",
      "1992       False             False             False              True   \n",
      "889        False             False             False              True   \n",
      "1362       False             False             False             False   \n",
      "\n",
      "      region_southwest  age_bin   age_bmi  \n",
      "1684             False      4.0  1.734561  \n",
      "862              False      1.0 -0.339153  \n",
      "1992             False      0.0 -1.055293  \n",
      "889              False      2.0  0.231109  \n",
      "1362              True      2.0  0.228540  \n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import OneHotEncoder, StandardScaler\n",
    "from sklearn.compose import ColumnTransformer\n",
    "from sklearn.pipeline import Pipeline\n",
    "\n",
    "# Загрузка данных\n",
    "df = pd.read_csv(\"..//static//csv//Medical_insurance.csv\")\n",
    "\n",
    "# Перемешивание данных\n",
    "df = df.sample(frac=1, random_state=42).reset_index(drop=True)\n",
    "\n",
    "# Разделение на признаки и целевую переменную\n",
    "X = df.drop('charges', axis=1)\n",
    "y = df['charges']\n",
    "\n",
    "# Разбиение на обучающую, контрольную и тестовую выборки\n",
    "X_train, X_temp, y_train, y_temp = train_test_split(X, y, test_size=0.3, random_state=42)\n",
    "X_val, X_test, y_val, y_test = train_test_split(X_temp, y_temp, test_size=0.5, random_state=42)\n",
    "\n",
    "# Определение категориальных и числовых признаков\n",
    "categorical_features = ['sex', 'smoker', 'region']\n",
    "numerical_features = ['age', 'bmi', 'children']\n",
    "\n",
    "# Унитарное кодирование категориальных признаков (one-hot encoding)\n",
    "train_data_encoded = pd.get_dummies(X_train, columns=categorical_features)\n",
    "val_data_encoded = pd.get_dummies(X_val, columns=categorical_features)\n",
    "test_data_encoded = pd.get_dummies(X_test, columns=categorical_features)\n",
    "\n",
    "# Дискретизация числовых признаков (пример для возраста)\n",
    "age_bins = [18, 30, 40, 50, 60, 100]\n",
    "train_data_encoded['age_bin'] = pd.cut(train_data_encoded['age'], bins=age_bins, labels=False)\n",
    "val_data_encoded['age_bin'] = pd.cut(val_data_encoded['age'], bins=age_bins, labels=False)\n",
    "test_data_encoded['age_bin'] = pd.cut(test_data_encoded['age'], bins=age_bins, labels=False)\n",
    "\n",
    "# «Ручной» синтез признаков (пример: комбинированный признак возраст и ИМТ)\n",
    "train_data_encoded['age_bmi'] = train_data_encoded['age'] * train_data_encoded['bmi']\n",
    "val_data_encoded['age_bmi'] = val_data_encoded['age'] * val_data_encoded['bmi']\n",
    "test_data_encoded['age_bmi'] = test_data_encoded['age'] * test_data_encoded['bmi']\n",
    "\n",
    "# Масштабирование числовых признаков\n",
    "numerical_features = ['age', 'bmi', 'children', 'age_bmi']\n",
    "scaler = StandardScaler()\n",
    "train_data_encoded[numerical_features] = scaler.fit_transform(train_data_encoded[numerical_features])\n",
    "val_data_encoded[numerical_features] = scaler.transform(val_data_encoded[numerical_features])\n",
    "test_data_encoded[numerical_features] = scaler.transform(test_data_encoded[numerical_features])\n",
    "\n",
    "# Вывод результатов\n",
    "print(\"Обучающая выборка после конструирования признаков:\")\n",
    "print(train_data_encoded.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Процесс конструирования признаков с применением фреймворка Featuretools:\n",
    "\n",
    "### 1. Оптимизация тарифов"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Обучающая выборка после конструирования признаков:\n",
      "            age       bmi  children  sex_female  sex_male  smoker_no  \\\n",
      "index                                                                  \n",
      "0      1.781292  0.374453 -0.907604        True     False       True   \n",
      "1     -0.083478 -0.570585 -0.085975        True     False       True   \n",
      "2     -1.087585 -0.495147 -0.907604        True     False       True   \n",
      "3      0.705463 -0.586335 -0.907604        True     False       True   \n",
      "4      0.777185 -0.680839 -0.907604       False      True       True   \n",
      "\n",
      "       smoker_yes  region_northeast  region_northwest  region_southeast  \\\n",
      "index                                                                     \n",
      "0           False             False              True             False   \n",
      "1           False              True             False             False   \n",
      "2           False             False             False              True   \n",
      "3           False             False             False              True   \n",
      "4           False             False             False             False   \n",
      "\n",
      "       region_southwest  age_bin   age_bmi  \n",
      "index                                       \n",
      "0                 False        4  1.734561  \n",
      "1                 False        1 -0.339153  \n",
      "2                 False        0 -1.055293  \n",
      "3                 False        2  0.231109  \n",
      "4                  True        2  0.228540  \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\midni\\AIM\\AIM-PIbd-32-Bulatova-K-R\\aimenv\\Lib\\site-packages\\featuretools\\entityset\\entityset.py:1733: UserWarning: index index not found in dataframe, creating new integer column\n",
      "  warnings.warn(\n",
      "c:\\Users\\midni\\AIM\\AIM-PIbd-32-Bulatova-K-R\\aimenv\\Lib\\site-packages\\featuretools\\synthesis\\deep_feature_synthesis.py:169: UserWarning: Only one dataframe in entityset, changing max_depth to 1 since deeper features cannot be created\n",
      "  warnings.warn(\n",
      "c:\\Users\\midni\\AIM\\AIM-PIbd-32-Bulatova-K-R\\aimenv\\Lib\\site-packages\\featuretools\\computational_backends\\feature_set_calculator.py:143: FutureWarning: The behavior of DataFrame concatenation with empty or all-NA entries is deprecated. In a future version, this will no longer exclude empty or all-NA columns when determining the result dtypes. To retain the old behavior, exclude the relevant entries before the concat operation.\n",
      "  df = pd.concat([df, default_df], sort=True)\n",
      "c:\\Users\\midni\\AIM\\AIM-PIbd-32-Bulatova-K-R\\aimenv\\Lib\\site-packages\\featuretools\\computational_backends\\feature_set_calculator.py:143: FutureWarning: The behavior of DataFrame concatenation with empty or all-NA entries is deprecated. In a future version, this will no longer exclude empty or all-NA columns when determining the result dtypes. To retain the old behavior, exclude the relevant entries before the concat operation.\n",
      "  df = pd.concat([df, default_df], sort=True)\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "import featuretools as ft\n",
    "\n",
    "# Загрузка данных\n",
    "df = pd.read_csv(\"..//static//csv//Medical_insurance.csv\")\n",
    "\n",
    "# Перемешивание данных\n",
    "df = df.sample(frac=1, random_state=42).reset_index(drop=True)\n",
    "\n",
    "# Разделение на признаки и целевую переменную\n",
    "X = df.drop('charges', axis=1)\n",
    "y = df['charges']\n",
    "\n",
    "# Разбиение на обучающую, контрольную и тестовую выборки\n",
    "X_train, X_temp, y_train, y_temp = train_test_split(X, y, test_size=0.3, random_state=42)\n",
    "X_val, X_test, y_val, y_test = train_test_split(X_temp, y_temp, test_size=0.5, random_state=42)\n",
    "\n",
    "# Определение категориальных и числовых признаков\n",
    "categorical_features = ['sex', 'smoker', 'region']\n",
    "numerical_features = ['age', 'bmi', 'children']\n",
    "\n",
    "# Унитарное кодирование категориальных признаков (one-hot encoding)\n",
    "X_train_encoded = pd.get_dummies(X_train, columns=categorical_features)\n",
    "X_val_encoded = pd.get_dummies(X_val, columns=categorical_features)\n",
    "X_test_encoded = pd.get_dummies(X_test, columns=categorical_features)\n",
    "\n",
    "# Дискретизация числовых признаков (пример для возраста)\n",
    "age_bins = [18, 30, 40, 50, 60, 100]\n",
    "X_train_encoded['age_bin'] = pd.cut(X_train_encoded['age'], bins=age_bins, labels=False)\n",
    "X_val_encoded['age_bin'] = pd.cut(X_val_encoded['age'], bins=age_bins, labels=False)\n",
    "X_test_encoded['age_bin'] = pd.cut(X_test_encoded['age'], bins=age_bins, labels=False)\n",
    "\n",
    "# «Ручной» синтез признаков (пример: комбинированный признак возраст и ИМТ)\n",
    "X_train_encoded['age_bmi'] = X_train_encoded['age'] * X_train_encoded['bmi']\n",
    "X_val_encoded['age_bmi'] = X_val_encoded['age'] * X_val_encoded['bmi']\n",
    "X_test_encoded['age_bmi'] = X_test_encoded['age'] * X_test_encoded['bmi']\n",
    "\n",
    "# Масштабирование числовых признаков\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "numerical_features = ['age', 'bmi', 'children', 'age_bmi']\n",
    "scaler = StandardScaler()\n",
    "X_train_encoded[numerical_features] = scaler.fit_transform(X_train_encoded[numerical_features])\n",
    "X_val_encoded[numerical_features] = scaler.transform(X_val_encoded[numerical_features])\n",
    "X_test_encoded[numerical_features] = scaler.transform(X_test_encoded[numerical_features])\n",
    "\n",
    "# Конструирование признаков с применением фреймворка Featuretools\n",
    "es = ft.EntitySet(id='insurance_data')\n",
    "es = es.add_dataframe(dataframe_name='train', dataframe=X_train_encoded, index='index')\n",
    "\n",
    "# Генерация признаков\n",
    "feature_matrix, feature_defs = ft.dfs(entityset=es, target_dataframe_name='train', max_depth=2)\n",
    "\n",
    "# Преобразование признаков для контрольной и тестовой выборок\n",
    "X_val_feature_matrix = ft.calculate_feature_matrix(features=feature_defs, entityset=es, instance_ids=X_val_encoded.index)\n",
    "X_test_feature_matrix = ft.calculate_feature_matrix(features=feature_defs, entityset=es, instance_ids=X_test_encoded.index)\n",
    "\n",
    "# Вывод результатов\n",
    "print(\"Обучающая выборка после конструирования признаков:\")\n",
    "print(feature_matrix.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2. Оценка рисков"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Обучающая выборка после конструирования признаков:\n",
      "            age       bmi  children  sex_female  sex_male  smoker_no  \\\n",
      "index                                                                  \n",
      "0      1.781292  0.374453 -0.907604        True     False       True   \n",
      "1     -0.083478 -0.570585 -0.085975        True     False       True   \n",
      "2     -1.087585 -0.495147 -0.907604        True     False       True   \n",
      "3      0.705463 -0.586335 -0.907604        True     False       True   \n",
      "4      0.777185 -0.680839 -0.907604       False      True       True   \n",
      "\n",
      "       smoker_yes  region_northeast  region_northwest  region_southeast  \\\n",
      "index                                                                     \n",
      "0           False             False              True             False   \n",
      "1           False              True             False             False   \n",
      "2           False             False             False              True   \n",
      "3           False             False             False              True   \n",
      "4           False             False             False             False   \n",
      "\n",
      "       region_southwest  age_bin   age_bmi  \n",
      "index                                       \n",
      "0                 False        4  1.734561  \n",
      "1                 False        1 -0.339153  \n",
      "2                 False        0 -1.055293  \n",
      "3                 False        2  0.231109  \n",
      "4                  True        2  0.228540  \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\midni\\AIM\\AIM-PIbd-32-Bulatova-K-R\\aimenv\\Lib\\site-packages\\featuretools\\entityset\\entityset.py:1733: UserWarning: index index not found in dataframe, creating new integer column\n",
      "  warnings.warn(\n",
      "c:\\Users\\midni\\AIM\\AIM-PIbd-32-Bulatova-K-R\\aimenv\\Lib\\site-packages\\featuretools\\synthesis\\deep_feature_synthesis.py:169: UserWarning: Only one dataframe in entityset, changing max_depth to 1 since deeper features cannot be created\n",
      "  warnings.warn(\n",
      "c:\\Users\\midni\\AIM\\AIM-PIbd-32-Bulatova-K-R\\aimenv\\Lib\\site-packages\\featuretools\\computational_backends\\feature_set_calculator.py:143: FutureWarning: The behavior of DataFrame concatenation with empty or all-NA entries is deprecated. In a future version, this will no longer exclude empty or all-NA columns when determining the result dtypes. To retain the old behavior, exclude the relevant entries before the concat operation.\n",
      "  df = pd.concat([df, default_df], sort=True)\n",
      "c:\\Users\\midni\\AIM\\AIM-PIbd-32-Bulatova-K-R\\aimenv\\Lib\\site-packages\\featuretools\\computational_backends\\feature_set_calculator.py:143: FutureWarning: The behavior of DataFrame concatenation with empty or all-NA entries is deprecated. In a future version, this will no longer exclude empty or all-NA columns when determining the result dtypes. To retain the old behavior, exclude the relevant entries before the concat operation.\n",
      "  df = pd.concat([df, default_df], sort=True)\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "import featuretools as ft\n",
    "\n",
    "# Загрузка данных\n",
    "df = pd.read_csv(\"..//static//csv//Medical_insurance.csv\")\n",
    "\n",
    "# Перемешивание данных\n",
    "df = df.sample(frac=1, random_state=42).reset_index(drop=True)\n",
    "\n",
    "# Разделение на признаки и целевую переменную\n",
    "X = df.drop('charges', axis=1)\n",
    "y = df['charges']\n",
    "\n",
    "# Разбиение на обучающую, контрольную и тестовую выборки\n",
    "X_train, X_temp, y_train, y_temp = train_test_split(X, y, test_size=0.3, random_state=42)\n",
    "X_val, X_test, y_val, y_test = train_test_split(X_temp, y_temp, test_size=0.5, random_state=42)\n",
    "\n",
    "# Определение категориальных и числовых признаков\n",
    "categorical_features = ['sex', 'smoker', 'region']\n",
    "numerical_features = ['age', 'bmi', 'children']\n",
    "\n",
    "# Унитарное кодирование категориальных признаков (one-hot encoding)\n",
    "X_train_encoded = pd.get_dummies(X_train, columns=categorical_features)\n",
    "X_val_encoded = pd.get_dummies(X_val, columns=categorical_features)\n",
    "X_test_encoded = pd.get_dummies(X_test, columns=categorical_features)\n",
    "\n",
    "# Дискретизация числовых признаков (пример для возраста)\n",
    "age_bins = [18, 30, 40, 50, 60, 100]\n",
    "X_train_encoded['age_bin'] = pd.cut(X_train_encoded['age'], bins=age_bins, labels=False)\n",
    "X_val_encoded['age_bin'] = pd.cut(X_val_encoded['age'], bins=age_bins, labels=False)\n",
    "X_test_encoded['age_bin'] = pd.cut(X_test_encoded['age'], bins=age_bins, labels=False)\n",
    "\n",
    "# «Ручной» синтез признаков (пример: комбинированный признак возраст и ИМТ)\n",
    "X_train_encoded['age_bmi'] = X_train_encoded['age'] * X_train_encoded['bmi']\n",
    "X_val_encoded['age_bmi'] = X_val_encoded['age'] * X_val_encoded['bmi']\n",
    "X_test_encoded['age_bmi'] = X_test_encoded['age'] * X_test_encoded['bmi']\n",
    "\n",
    "# Масштабирование числовых признаков\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "\n",
    "numerical_features = ['age', 'bmi', 'children', 'age_bmi']\n",
    "scaler = StandardScaler()\n",
    "X_train_encoded[numerical_features] = scaler.fit_transform(X_train_encoded[numerical_features])\n",
    "X_val_encoded[numerical_features] = scaler.transform(X_val_encoded[numerical_features])\n",
    "X_test_encoded[numerical_features] = scaler.transform(X_test_encoded[numerical_features])\n",
    "\n",
    "# Конструирование признаков с применением фреймворка Featuretools\n",
    "es = ft.EntitySet(id='insurance_data')\n",
    "es = es.add_dataframe(dataframe_name='train', dataframe=X_train_encoded, index='index')\n",
    "\n",
    "# Генерация признаков\n",
    "feature_matrix, feature_defs = ft.dfs(entityset=es, target_dataframe_name='train', max_depth=2)\n",
    "\n",
    "# Преобразование признаков для контрольной и тестовой выборок\n",
    "X_val_feature_matrix = ft.calculate_feature_matrix(features=feature_defs, entityset=es, instance_ids=X_val_encoded.index)\n",
    "X_test_feature_matrix = ft.calculate_feature_matrix(features=feature_defs, entityset=es, instance_ids=X_test_encoded.index)\n",
    "\n",
    "# Вывод результатов\n",
    "print(\"Обучающая выборка после конструирования признаков:\")\n",
    "print(feature_matrix.head())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Оценка качества наборов признаков по критериям:\n",
    "\n",
    "### 1. Предсказательная способность\n",
    "\n",
    "**Определение:** \n",
    "Предсказательная способность набора признаков определяет, насколько хорошо эти признаки позволяют модели предсказывать целевую переменную.\n",
    "\n",
    "**Оценка:**\n",
    "- **Обучающая выборка:** \n",
    "  - Оценивается с помощью метрик качества модели (например, RMSE, MAE, R²) на обучающей выборке.\n",
    "  - Высокие значения метрик указывают на высокую предсказательную способность.\n",
    "- **Контрольная и тестовая выборки:**\n",
    "  - Оценивается с помощью метрик качества модели на контрольной и тестовой выборках.\n",
    "  - Близкие значения метрик на всех выборках указывают на хорошую обобщающую способность модели.\n",
    "\n",
    "### 2. Скорость вычисления\n",
    "\n",
    "**Определение:**\n",
    "Скорость вычисления набора признаков определяет, насколько быстро можно вычислить эти признаки и обучить модель на них.\n",
    "\n",
    "**Оценка:**\n",
    "- **Время вычисления признаков:**\n",
    "  - Измеряется время, затрачиваемое на вычисление признаков для всех выборок.\n",
    "  - Меньшее время указывает на более быстрое вычисление.\n",
    "- **Время обучения модели:**\n",
    "  - Измеряется время, затрачиваемое на обучение модели на вычисленных признаках.\n",
    "  - Меньшее время указывает на более быстрое обучение.\n",
    "\n",
    "### 3. Надежность\n",
    "\n",
    "**Определение:**\n",
    "Надежность набора признаков определяет, насколько стабильно модель, обученная на этих признаках, показывает хорошие результаты на разных выборках данных.\n",
    "\n",
    "**Оценка:**\n",
    "- **Стабильность метрик:**\n",
    "  - Оценивается стабильность метрик качества модели (например, RMSE, MAE, R²) на разных выборках данных.\n",
    "  - Близкие значения метрик на разных выборках указывают на высокую надежность.\n",
    "- **Устойчивость к изменениям данных:**\n",
    "  - Оценивается, как меняются метрики качества модели при добавлении или удалении данных.\n",
    "  - Небольшие изменения метрик указывают на высокую устойчивость.\n",
    "\n",
    "### 4. Корреляция\n",
    "\n",
    "**Определение:**\n",
    "Корреляция набора признаков определяет, насколько сильно признаки коррелируют друг с другом и с целевой переменной.\n",
    "\n",
    "**Оценка:**\n",
    "- **Корреляция между признаками:**\n",
    "  - Оценивается с помощью матрицы корреляции признаков.\n",
    "  - Высокая корреляция между признаками может привести к мультиколлинеарности, что снижает качество модели.\n",
    "- **Корреляция с целевой переменной:**\n",
    "  - Оценивается с помощью коэффициента корреляции Пирсона или Спирмена.\n",
    "  - Высокая корреляция с целевой переменной указывает на высокую предсказательную способность признаков.\n",
    "\n",
    "### 5. Цельность\n",
    "\n",
    "**Определение:**\n",
    "Цельность набора признаков определяет, насколько хорошо признаки соответствуют бизнес-целям и задачам модели.\n",
    "\n",
    "**Оценка:**\n",
    "- **Соответствие бизнес-целям:**\n",
    "  - Оценивается, насколько признаки помогают решать поставленные бизнес-задачи (например, оптимизация тарифов, оценка рисков).\n",
    "  - Признаки, которые помогают решать бизнес-задачи, считаются целесообразными.\n",
    "- **Интерпретируемость:**\n",
    "  - Оценивается, насколько легко интерпретировать значения признаков и их влияние на целевую переменную.\n",
    "  - Интерпретируемые признаки считаются более целесообразными."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\midni\\AIM\\AIM-PIbd-32-Bulatova-K-R\\aimenv\\Lib\\site-packages\\sklearn\\metrics\\_regression.py:492: FutureWarning: 'squared' is deprecated in version 1.4 and will be removed in 1.6. To calculate the root mean squared error, use the function'root_mean_squared_error'.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "RMSE: 2750.642231395856\n",
      "R²: 0.9507037692209687\n",
      "MAE: 1279.1669853384874\n",
      "Cross-validated RMSE: 3242.964333689781\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train RMSE: 1072.3827255198853\n",
      "Train R²: 0.9921277274068127\n",
      "Train MAE: 480.25126389741285\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "c:\\Users\\midni\\AIM\\AIM-PIbd-32-Bulatova-K-R\\aimenv\\Lib\\site-packages\\sklearn\\metrics\\_regression.py:492: FutureWarning: 'squared' is deprecated in version 1.4 and will be removed in 1.6. To calculate the root mean squared error, use the function'root_mean_squared_error'.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1000x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.metrics import mean_squared_error, r2_score, mean_absolute_error\n",
    "from sklearn.model_selection import cross_val_score\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "# Загрузка данных\n",
    "df = pd.read_csv(\"..//static//csv//Medical_insurance.csv\")\n",
    "\n",
    "# Предобработка данных\n",
    "# Преобразуем категориальные переменные в числовые\n",
    "df = pd.get_dummies(df, drop_first=True)\n",
    "\n",
    "# Разделение данных на признаки и целевую переменную\n",
    "X = df.drop('charges', axis=1)\n",
    "y = df['charges']\n",
    "\n",
    "# Разделение данных на обучающую и тестовую выборки\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)\n",
    "\n",
    "# Выбор модели\n",
    "model = RandomForestRegressor(random_state=42)\n",
    "\n",
    "# Обучение модели\n",
    "model.fit(X_train, y_train)\n",
    "\n",
    "# Предсказание и оценка\n",
    "y_pred = model.predict(X_test)\n",
    "\n",
    "rmse = mean_squared_error(y_test, y_pred, squared=False)\n",
    "r2 = r2_score(y_test, y_pred)\n",
    "mae = mean_absolute_error(y_test, y_pred)\n",
    "\n",
    "print(f\"RMSE: {rmse}\")\n",
    "print(f\"R²: {r2}\")\n",
    "print(f\"MAE: {mae}\")\n",
    "\n",
    "# Кросс-валидация\n",
    "scores = cross_val_score(model, X_train, y_train, cv=5, scoring='neg_mean_squared_error')\n",
    "rmse_cv = (-scores.mean())**0.5\n",
    "print(f\"Cross-validated RMSE: {rmse_cv}\")\n",
    "\n",
    "# Анализ важности признаков\n",
    "feature_importances = model.feature_importances_\n",
    "feature_names = X_train.columns\n",
    "\n",
    "importance_df = pd.DataFrame({'Feature': feature_names, 'Importance': feature_importances})\n",
    "importance_df = importance_df.sort_values(by='Importance', ascending=False)\n",
    "\n",
    "plt.figure(figsize=(10, 6))\n",
    "sns.barplot(x='Importance', y='Feature', data=importance_df)\n",
    "plt.title('Feature Importance')\n",
    "plt.show()\n",
    "\n",
    "# Проверка на переобучение\n",
    "y_train_pred = model.predict(X_train)\n",
    "\n",
    "rmse_train = mean_squared_error(y_train, y_train_pred, squared=False)\n",
    "r2_train = r2_score(y_train, y_train_pred)\n",
    "mae_train = mean_absolute_error(y_train, y_train_pred)\n",
    "\n",
    "print(f\"Train RMSE: {rmse_train}\")\n",
    "print(f\"Train R²: {r2_train}\")\n",
    "print(f\"Train MAE: {mae_train}\")\n",
    "\n",
    "# Визуализация результатов\n",
    "plt.figure(figsize=(10, 6))\n",
    "plt.scatter(y_test, y_pred, alpha=0.5)\n",
    "plt.plot([y_test.min(), y_test.max()], [y_test.min(), y_test.max()], 'k--', lw=2)\n",
    "plt.xlabel('Actual Charges')\n",
    "plt.ylabel('Predicted Charges')\n",
    "plt.title('Actual vs Predicted Charges')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Вывод по данным:\n",
    "\n",
    "1. Train RMSE: Значение RMSE на обучающей выборке составляет 1072.38, что указывает на среднеквадратичную ошибку в предсказании стоимости медицинского страхования.\n",
    "\n",
    "2. Train R²: Значение R² на обучающей выборке составляет 0.9921, что говорит о том, что модель объясняет 99.21% вариации в данных. Это очень высокий показатель, что может указывать на потенциальное переобучение.\n",
    "\n",
    "3. Train MAE: Значение MAE на обучающей выборке составляет 480.25, что указывает на среднюю абсолютную ошибку в предсказании стоимости медицинского страхования."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "aimenv",
   "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": "3.12.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}