{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Данные успешно сохранены в processed_data.npz\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "def task1_preprocessing() -> None:\n",
    "    # 1. Загрузка данных\n",
    "    try:\n",
    "        data = pd.read_csv(\"dataset/opsd_raw.csv\", parse_dates=[\"utc_timestamp\"])\n",
    "    except FileNotFoundError:\n",
    "        print(\"Ошибка: Файл dataset/opsd_raw.csv не найден.\")\n",
    "        return\n",
    "    \n",
    "    # 2. Выбор релевантных столбцов для Дании\n",
    "    cols = [\n",
    "        \"utc_timestamp\",\n",
    "        \"DK_load_actual_entsoe_transparency\",\n",
    "        \"DK_wind_generation_actual\", \n",
    "        \"DK_solar_generation_actual\"\n",
    "    ]\n",
    "    # Проверка наличия столбцов в данных\n",
    "    missing_cols = [col for col in cols if col not in data.columns]\n",
    "    if missing_cols:\n",
    "        print(f\"Ошибка: Отсутствуют столбцы {missing_cols} в данных.\")\n",
    "        return\n",
    "    data = data[cols].copy()\n",
    "    data.columns = [\"timestamp\", \"load\", \"wind\", \"solar\"]\n",
    "    \n",
    "    # 3. Обработка пропущенных значений\n",
    "    for col in [\"load\", \"wind\", \"solar\"]:\n",
    "        data[col].fillna(method='ffill', inplace=True)  # Заполняем пропуски предыдущими значениями\n",
    "        data[col].fillna(method='bfill', inplace=True)  # Если остались пропуски, заполняем следующими значениями\n",
    "\n",
    "    # 4. Формирование массивов 24x3 (часы × признаки)\n",
    "    data.set_index(\"timestamp\", inplace=True)\n",
    "    \n",
    "    # Группируем по дням и оставляем только дни с ровно 24 записями\n",
    "    daily_data = (\n",
    "        data.groupby(data.index.date)  # Группировка по дате\n",
    "        .filter(lambda x: len(x) == 24)  # Фильтруем группы с ровно 24 строками\n",
    "        .groupby(data.index.date)  # Повторная группировка по дате\n",
    "        .apply(lambda x: x[[\"load\", \"wind\", \"solar\"]].values.T)  # Преобразуем в массивы\n",
    "    )\n",
    "    \n",
    "    # Преобразуем индексы в формат datetime для дальнейшей работы\n",
    "    daily_data.index = pd.to_datetime(daily_data.index)\n",
    "    \n",
    "    # 5. Создание DataFrame с массивами и метками сезонов\n",
    "    df = pd.DataFrame({\n",
    "        \"timestamp\": daily_data.index,\n",
    "        \"data\": daily_data.values,\n",
    "        \"season\": daily_data.index.map(lambda x: get_season(x.month))\n",
    "    })\n",
    "    \n",
    "    # 6. Разделение данных\n",
    "    X = np.stack(df[\"data\"].values)\n",
    "    y = df[\"season\"].values\n",
    "    \n",
    "    X_train, X_temp, y_train, y_temp = train_test_split(\n",
    "        X, y, test_size=0.3, stratify=y, random_state=42\n",
    "    )\n",
    "    X_val, X_test, y_val, y_test = train_test_split(\n",
    "        X_temp, y_temp, test_size=0.5, stratify=y_temp, random_state=42\n",
    "    )\n",
    "    \n",
    "    # 7. Масштабирование\n",
    "    scaler = StandardScaler()\n",
    "    X_train_scaled = scaler.fit_transform(X_train.reshape(-1, 3)).reshape(X_train.shape)\n",
    "    X_val_scaled = scaler.transform(X_val.reshape(-1, 3)).reshape(X_val.shape)\n",
    "    X_test_scaled = scaler.transform(X_test.reshape(-1, 3)).reshape(X_test.shape)\n",
    "    \n",
    "    # 8. Сохранение данных\n",
    "    try:\n",
    "        np.savez(\"processed_data.npz\",\n",
    "                X_train=X_train_scaled,\n",
    "                X_val=X_val_scaled,\n",
    "                X_test=X_test_scaled,\n",
    "                y_train=y_train,\n",
    "                y_val=y_val,\n",
    "                y_test=y_test)\n",
    "        print(\"Данные успешно сохранены в processed_data.npz\")\n",
    "    except Exception as e:\n",
    "        print(f\"Ошибка при сохранении данных: {e}\")\n",
    "        return\n",
    "    \n",
    "    # 9. Визуализация\n",
    "    plot_sample_profiles(df)\n",
    "\n",
    "def get_season(month: int) -> str:\n",
    "    if month in [12, 1, 2]: return \"winter\"\n",
    "    if month in [3, 4, 5]: return \"spring\"\n",
    "    if month in [6, 7, 8]: return \"summer\"\n",
    "    return \"autumn\"\n",
    "\n",
    "def plot_sample_profiles(df: pd.DataFrame) -> None:\n",
    "    plt.figure(figsize=(12, 6))\n",
    "    for season in [\"winter\", \"spring\", \"summer\", \"autumn\"]:\n",
    "        sample = df[df[\"season\"] == season].iloc[0][\"data\"]\n",
    "        plt.plot(sample[0], label=f\"{season} load\")\n",
    "    plt.title(\"Суточные профили потребления энергии по сезонам\")\n",
    "    plt.xlabel(\"Час дня (0-23)\")\n",
    "    plt.ylabel(\"Мощность (МВт)\")\n",
    "    plt.legend()\n",
    "    plt.grid(True)\n",
    "    plt.savefig(\"load_profiles.png\")\n",
    "    plt.show()\n",
    "\n",
    "# Запуск preprocessing\n",
    "task1_preprocessing()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Запуск GridSearchCV...\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n",
      "/var/data/python/lib/python3.12/site-packages/sklearn/neural_network/_multilayer_perceptron.py:691: ConvergenceWarning: Stochastic Optimizer: Maximum iterations (500) reached and the optimization hasn't converged yet.\n",
      "  warnings.warn(\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Лучшие параметры: {'activation': 'tanh', 'alpha': 0.0001, 'hidden_layer_sizes': (100, 100), 'solver': 'adam'}\n",
      "Точность (Accuracy): 0.7810\n",
      "Точность (Precision): 0.7951\n",
      "Полнота (Recall): 0.7810\n",
      "F1-Score: 0.7708\n"
     ]
    }
   ],
   "source": [
    "import numpy as np\n",
    "from sklearn.model_selection import train_test_split, GridSearchCV\n",
    "from sklearn.neural_network import MLPClassifier\n",
    "from sklearn.metrics import accuracy_score, precision_score, recall_score, f1_score\n",
    "import warnings\n",
    "\n",
    "def task2_model_training() -> None:\n",
    "    # Игнорируем предупреждения для более чистого вывода\n",
    "    warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "    # 1. Загрузка предварительно обработанных данных\n",
    "    try:\n",
    "        data = np.load(\"processed_data.npz\", allow_pickle=True)\n",
    "        X_train, X_val, X_test = data[\"X_train\"], data[\"X_val\"], data[\"X_test\"]\n",
    "        y_train, y_val, y_test = data[\"y_train\"], data[\"y_val\"], data[\"y_test\"]\n",
    "    except FileNotFoundError:\n",
    "        print(\"Ошибка: Файл processed_data.npz не найден. Запустите task1_preprocessing() перед выполнением этой задачи.\")\n",
    "        return\n",
    "    \n",
    "    # 2. Определение модели MLPClassifier\n",
    "    mlp = MLPClassifier(max_iter=500, random_state=42)\n",
    "    \n",
    "    # 3. Настройка гиперпараметров с использованием GridSearchCV\n",
    "    param_grid = {\n",
    "        \"hidden_layer_sizes\": [(50,), (100,), (50, 50), (100, 100)],\n",
    "        \"activation\": [\"relu\", \"tanh\"],\n",
    "        \"solver\": [\"adam\", \"sgd\"],\n",
    "        \"alpha\": [0.0001, 0.001],\n",
    "    }\n",
    "    grid_search = GridSearchCV(mlp, param_grid, cv=5, scoring=\"accuracy\", n_jobs=-1)\n",
    "    \n",
    "    print(\"Запуск GridSearchCV...\")\n",
    "    # Убедитесь, что X_train не содержит NaN или бесконечные значения\n",
    "    if np.isnan(X_train).any() or not np.isfinite(X_train).all():\n",
    "        print(\"Ошибка: X_train содержит NaN или бесконечные значения. Проверьте данные.\")\n",
    "        return\n",
    "    \n",
    "    grid_search.fit(X_train.reshape(X_train.shape[0], -1), y_train)\n",
    "    \n",
    "    # 4. Лучшая модель и её параметры\n",
    "    best_model = grid_search.best_estimator_\n",
    "    print(\"Лучшие параметры:\", grid_search.best_params_)\n",
    "    \n",
    "    # 5. Оценка модели на тестовом наборе\n",
    "    # Убедитесь, что X_test не содержит NaN или бесконечные значения\n",
    "    if np.isnan(X_test).any() or not np.isfinite(X_test).all():\n",
    "        print(\"Ошибка: X_test содержит NaN или бесконечные значения. Проверьте данные.\")\n",
    "        return\n",
    "    \n",
    "    y_pred = best_model.predict(X_test.reshape(X_test.shape[0], -1))\n",
    "    \n",
    "    accuracy = accuracy_score(y_test, y_pred)\n",
    "    precision = precision_score(y_test, y_pred, average=\"weighted\")\n",
    "    recall = recall_score(y_test, y_pred, average=\"weighted\")\n",
    "    f1 = f1_score(y_test, y_pred, average=\"weighted\")\n",
    "    \n",
    "    print(f\"Точность (Accuracy): {accuracy:.4f}\")\n",
    "    print(f\"Точность (Precision): {precision:.4f}\")\n",
    "    print(f\"Полнота (Recall): {recall:.4f}\")\n",
    "    print(f\"F1-Score: {f1:.4f}\")\n",
    "\n",
    "# Запуск обучения модели\n",
    "task2_model_training()\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}