{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "c8941fb1",
   "metadata": {},
   "source": [
    "\n",
    "# Снижение размерности и регрессия по данным London House Price\n",
    "\n",
    "Ноутбук оптимизирован по памяти:\n",
    "- используется только числовые независимые признаки\n",
    "- пропуски заполняются медианой\n",
    "- сравниваются только методы, подходящие для большого датасета\n",
    "- без KernelPCA, так как он требует квадратичную матрицу по числу объектов\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "39dbbf71",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "import warnings\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.preprocessing import StandardScaler\n",
    "from sklearn.decomposition import PCA, TruncatedSVD, FactorAnalysis\n",
    "from sklearn.impute import SimpleImputer\n",
    "from sklearn.linear_model import Ridge\n",
    "from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "ee2818d7",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Размер датасета: (418201, 28)\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>fullAddress</th>\n",
       "      <th>postcode</th>\n",
       "      <th>country</th>\n",
       "      <th>outcode</th>\n",
       "      <th>latitude</th>\n",
       "      <th>longitude</th>\n",
       "      <th>bathrooms</th>\n",
       "      <th>bedrooms</th>\n",
       "      <th>floorAreaSqM</th>\n",
       "      <th>livingRooms</th>\n",
       "      <th>...</th>\n",
       "      <th>saleEstimate_upperPrice</th>\n",
       "      <th>saleEstimate_confidenceLevel</th>\n",
       "      <th>saleEstimate_ingestedAt</th>\n",
       "      <th>saleEstimate_valueChange.numericChange</th>\n",
       "      <th>saleEstimate_valueChange.percentageChange</th>\n",
       "      <th>saleEstimate_valueChange.saleDate</th>\n",
       "      <th>history_date</th>\n",
       "      <th>history_price</th>\n",
       "      <th>history_percentageChange</th>\n",
       "      <th>history_numericChange</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Flat 9, 35 Furnival Street, London, EC4A 1JQ</td>\n",
       "      <td>EC4A 1JQ</td>\n",
       "      <td>England</td>\n",
       "      <td>EC4A</td>\n",
       "      <td>51.517282</td>\n",
       "      <td>-0.110314</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>45.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>630000.0</td>\n",
       "      <td>HIGH</td>\n",
       "      <td>2024-10-07T13:26:59.894Z</td>\n",
       "      <td>244000.0</td>\n",
       "      <td>68.539326</td>\n",
       "      <td>2010-03-30</td>\n",
       "      <td>1995-01-02</td>\n",
       "      <td>830000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Flat 6, 35 Furnival Street, London, EC4A 1JQ</td>\n",
       "      <td>EC4A 1JQ</td>\n",
       "      <td>England</td>\n",
       "      <td>EC4A</td>\n",
       "      <td>51.517282</td>\n",
       "      <td>-0.110314</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>660000.0</td>\n",
       "      <td>MEDIUM</td>\n",
       "      <td>2024-10-07T13:26:59.894Z</td>\n",
       "      <td>425000.0</td>\n",
       "      <td>242.857143</td>\n",
       "      <td>2000-05-26</td>\n",
       "      <td>1995-01-02</td>\n",
       "      <td>830000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Flat 35, Octavia House, Medway Street, London,...</td>\n",
       "      <td>SW1P 2TA</td>\n",
       "      <td>England</td>\n",
       "      <td>SW1P</td>\n",
       "      <td>51.495505</td>\n",
       "      <td>-0.132379</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>71.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>834000.0</td>\n",
       "      <td>MEDIUM</td>\n",
       "      <td>2025-01-10T11:04:57.114Z</td>\n",
       "      <td>49000.0</td>\n",
       "      <td>6.901408</td>\n",
       "      <td>2019-09-04</td>\n",
       "      <td>1995-01-03</td>\n",
       "      <td>249950</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>24 Chester Court, Lomond Grove, London, SE5 7HS</td>\n",
       "      <td>SE5 7HS</td>\n",
       "      <td>England</td>\n",
       "      <td>SE5</td>\n",
       "      <td>51.478185</td>\n",
       "      <td>-0.092201</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>64.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>407000.0</td>\n",
       "      <td>HIGH</td>\n",
       "      <td>2024-10-07T13:26:59.894Z</td>\n",
       "      <td>28000.0</td>\n",
       "      <td>7.777778</td>\n",
       "      <td>2024-01-25</td>\n",
       "      <td>1995-01-03</td>\n",
       "      <td>32000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>18 Alexandra Gardens, London, N10 3RL</td>\n",
       "      <td>N10 3RL</td>\n",
       "      <td>England</td>\n",
       "      <td>N10</td>\n",
       "      <td>51.588774</td>\n",
       "      <td>-0.139599</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>137.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>1324000.0</td>\n",
       "      <td>HIGH</td>\n",
       "      <td>2024-10-07T13:26:59.894Z</td>\n",
       "      <td>81000.0</td>\n",
       "      <td>6.864407</td>\n",
       "      <td>2022-12-14</td>\n",
       "      <td>1995-01-03</td>\n",
       "      <td>133000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                         fullAddress  postcode  country  \\\n",
       "0       Flat 9, 35 Furnival Street, London, EC4A 1JQ  EC4A 1JQ  England   \n",
       "1       Flat 6, 35 Furnival Street, London, EC4A 1JQ  EC4A 1JQ  England   \n",
       "2  Flat 35, Octavia House, Medway Street, London,...  SW1P 2TA  England   \n",
       "3    24 Chester Court, Lomond Grove, London, SE5 7HS   SE5 7HS  England   \n",
       "4              18 Alexandra Gardens, London, N10 3RL   N10 3RL  England   \n",
       "\n",
       "  outcode   latitude  longitude  bathrooms  bedrooms  floorAreaSqM  \\\n",
       "0    EC4A  51.517282  -0.110314        1.0       1.0          45.0   \n",
       "1    EC4A  51.517282  -0.110314        NaN       NaN           NaN   \n",
       "2    SW1P  51.495505  -0.132379        2.0       2.0          71.0   \n",
       "3     SE5  51.478185  -0.092201        1.0       1.0          64.0   \n",
       "4     N10  51.588774  -0.139599        1.0       4.0         137.0   \n",
       "\n",
       "   livingRooms  ... saleEstimate_upperPrice saleEstimate_confidenceLevel  \\\n",
       "0          1.0  ...                630000.0                         HIGH   \n",
       "1          NaN  ...                660000.0                       MEDIUM   \n",
       "2          1.0  ...                834000.0                       MEDIUM   \n",
       "3          1.0  ...                407000.0                         HIGH   \n",
       "4          2.0  ...               1324000.0                         HIGH   \n",
       "\n",
       "    saleEstimate_ingestedAt  saleEstimate_valueChange.numericChange  \\\n",
       "0  2024-10-07T13:26:59.894Z                                244000.0   \n",
       "1  2024-10-07T13:26:59.894Z                                425000.0   \n",
       "2  2025-01-10T11:04:57.114Z                                 49000.0   \n",
       "3  2024-10-07T13:26:59.894Z                                 28000.0   \n",
       "4  2024-10-07T13:26:59.894Z                                 81000.0   \n",
       "\n",
       "   saleEstimate_valueChange.percentageChange  \\\n",
       "0                                  68.539326   \n",
       "1                                 242.857143   \n",
       "2                                   6.901408   \n",
       "3                                   7.777778   \n",
       "4                                   6.864407   \n",
       "\n",
       "   saleEstimate_valueChange.saleDate  history_date  history_price  \\\n",
       "0                         2010-03-30    1995-01-02         830000   \n",
       "1                         2000-05-26    1995-01-02         830000   \n",
       "2                         2019-09-04    1995-01-03         249950   \n",
       "3                         2024-01-25    1995-01-03          32000   \n",
       "4                         2022-12-14    1995-01-03         133000   \n",
       "\n",
       "   history_percentageChange history_numericChange  \n",
       "0                       NaN                   NaN  \n",
       "1                       NaN                   NaN  \n",
       "2                       NaN                   NaN  \n",
       "3                       NaN                   NaN  \n",
       "4                       NaN                   NaN  \n",
       "\n",
       "[5 rows x 28 columns]"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "file_path = \"kaggle_london_house_price_data.csv\"  # поменяй имя при необходимости\n",
    "\n",
    "df = pd.read_csv(file_path, low_memory=False)\n",
    "df.columns = df.columns.str.strip()\n",
    "\n",
    "print(\"Размер датасета:\", df.shape)\n",
    "display(df.head())\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6673e5bc",
   "metadata": {},
   "source": [
    "## Целевая переменная"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "a8cb98ab",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "saleEstimate_currentPrice\n"
     ]
    }
   ],
   "source": [
    "\n",
    "target = \"saleEstimate_currentPrice\"\n",
    "print(target)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "0ad78b17",
   "metadata": {},
   "source": [
    "## Отбор независимых числовых признаков"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "a13ad267",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Число признаков: 14\n",
      "Размер выборки: (417561, 14)\n"
     ]
    }
   ],
   "source": [
    "\n",
    "numeric_cols = df.select_dtypes(include=[\"number\"]).columns.tolist()\n",
    "\n",
    "exclude_cols = [\n",
    "    target,\n",
    "    \"saleEstimate_lowerPrice\",\n",
    "    \"saleEstimate_upperPrice\"\n",
    "]\n",
    "\n",
    "feature_cols = [c for c in numeric_cols if c not in exclude_cols]\n",
    "\n",
    "data = df[feature_cols + [target]].copy()\n",
    "data = data.dropna(axis=1, how=\"all\")\n",
    "data = data.dropna(subset=[target]).copy()\n",
    "\n",
    "feature_cols = [c for c in data.columns if c != target]\n",
    "\n",
    "X = data[feature_cols].copy()\n",
    "y = data[target].copy()\n",
    "\n",
    "imputer = SimpleImputer(strategy=\"median\")\n",
    "X_imp = pd.DataFrame(imputer.fit_transform(X), columns=feature_cols, index=X.index)\n",
    "\n",
    "stds = X_imp.std()\n",
    "keep_cols = stds[stds > 0].index.tolist()\n",
    "X_imp = X_imp[keep_cols]\n",
    "\n",
    "print(\"Число признаков:\", len(X_imp.columns))\n",
    "print(\"Размер выборки:\", X_imp.shape)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "81027349",
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "scaler = StandardScaler()\n",
    "X_scaled = scaler.fit_transform(X_imp)\n",
    "\n",
    "idx = np.arange(len(X_imp))\n",
    "idx_train, idx_test = train_test_split(idx, test_size=0.2, random_state=42)\n",
    "\n",
    "y_train = y.iloc[idx_train]\n",
    "y_test = y.iloc[idx_test]\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e3692ee2",
   "metadata": {},
   "source": [
    "## Снижение размерности несколькими способами"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "a6fccc81",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Method</th>\n",
       "      <th>Explained_variance</th>\n",
       "      <th>MAE_test</th>\n",
       "      <th>RMSE_test</th>\n",
       "      <th>R2_test</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>FactorAnalysis</td>\n",
       "      <td>NaN</td>\n",
       "      <td>96334.480523</td>\n",
       "      <td>184830.641094</td>\n",
       "      <td>0.960118</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>TruncatedSVD</td>\n",
       "      <td>0.48772</td>\n",
       "      <td>215160.216089</td>\n",
       "      <td>363857.639474</td>\n",
       "      <td>0.845443</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>PCA</td>\n",
       "      <td>0.48772</td>\n",
       "      <td>215160.216089</td>\n",
       "      <td>363857.639474</td>\n",
       "      <td>0.845443</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           Method  Explained_variance       MAE_test      RMSE_test   R2_test\n",
       "0  FactorAnalysis                 NaN   96334.480523  184830.641094  0.960118\n",
       "1    TruncatedSVD             0.48772  215160.216089  363857.639474  0.845443\n",
       "2             PCA             0.48772  215160.216089  363857.639474  0.845443"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "results = []\n",
    "reduced_data = {}\n",
    "\n",
    "def eval_regression(X_red, method_name, explained_variance=np.nan):\n",
    "    Xr = pd.DataFrame(X_red, columns=[\"C1\", \"C2\"], index=X_imp.index)\n",
    "    X_train = Xr.iloc[idx_train]\n",
    "    X_test = Xr.iloc[idx_test]\n",
    "\n",
    "    model = Ridge(alpha=1.0)\n",
    "    model.fit(X_train, y_train)\n",
    "    pred = model.predict(X_test)\n",
    "\n",
    "    results.append({\n",
    "        \"Method\": method_name,\n",
    "        \"Explained_variance\": explained_variance,\n",
    "        \"MAE_test\": mean_absolute_error(y_test, pred),\n",
    "        \"RMSE_test\": mean_squared_error(y_test, pred) ** 0.5,\n",
    "        \"R2_test\": r2_score(y_test, pred)\n",
    "    })\n",
    "    reduced_data[method_name] = Xr\n",
    "\n",
    "# PCA\n",
    "pca = PCA(n_components=2, random_state=42)\n",
    "X_pca = pca.fit_transform(X_scaled)\n",
    "eval_regression(X_pca, \"PCA\", pca.explained_variance_ratio_.sum())\n",
    "\n",
    "# Truncated SVD\n",
    "svd = TruncatedSVD(n_components=2, random_state=42)\n",
    "X_svd = svd.fit_transform(X_scaled)\n",
    "eval_regression(X_svd, \"TruncatedSVD\", svd.explained_variance_ratio_.sum())\n",
    "\n",
    "# Factor Analysis\n",
    "fa = FactorAnalysis(n_components=2, random_state=42)\n",
    "X_fa = fa.fit_transform(X_scaled)\n",
    "eval_regression(X_fa, \"FactorAnalysis\", np.nan)\n",
    "\n",
    "results_df = pd.DataFrame(results).sort_values([\"R2_test\", \"RMSE_test\"], ascending=[False, True]).reset_index(drop=True)\n",
    "display(results_df)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "34ccaeba",
   "metadata": {},
   "source": [
    "## Лучший метод"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "2c98ab8e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Лучший метод: FactorAnalysis\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Comp1</th>\n",
       "      <th>Comp2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-0.369367</td>\n",
       "      <td>0.383211</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>-0.369367</td>\n",
       "      <td>0.378143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>-0.197139</td>\n",
       "      <td>-0.217205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>-0.469856</td>\n",
       "      <td>-0.242659</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0.348269</td>\n",
       "      <td>-0.648269</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Comp1     Comp2\n",
       "0 -0.369367  0.383211\n",
       "1 -0.369367  0.378143\n",
       "2 -0.197139 -0.217205\n",
       "3 -0.469856 -0.242659\n",
       "4  0.348269 -0.648269"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "best_method = results_df.loc[0, \"Method\"]\n",
    "print(\"Лучший метод:\", best_method)\n",
    "\n",
    "X_best = reduced_data[best_method].copy()\n",
    "X_best.columns = [\"Comp1\", \"Comp2\"]\n",
    "display(X_best.head())\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e5fb58c9",
   "metadata": {},
   "source": [
    "## Регрессия по новым признакам"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "22039f28",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Метрика</th>\n",
       "      <th>Значение</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>MAE_train</td>\n",
       "      <td>95678.679653</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>RMSE_train</td>\n",
       "      <td>180566.864840</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>R2_train</td>\n",
       "      <td>0.961393</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>MAE_test</td>\n",
       "      <td>96334.480523</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>RMSE_test</td>\n",
       "      <td>184830.641094</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>R2_test</td>\n",
       "      <td>0.960118</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Метрика       Значение\n",
       "0   MAE_train   95678.679653\n",
       "1  RMSE_train  180566.864840\n",
       "2    R2_train       0.961393\n",
       "3    MAE_test   96334.480523\n",
       "4   RMSE_test  184830.641094\n",
       "5     R2_test       0.960118"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "X_train = X_best.iloc[idx_train]\n",
    "X_test = X_best.iloc[idx_test]\n",
    "\n",
    "model_final = Ridge(alpha=1.0)\n",
    "model_final.fit(X_train, y_train)\n",
    "\n",
    "pred_train = model_final.predict(X_train)\n",
    "pred_test = model_final.predict(X_test)\n",
    "\n",
    "metrics_final = pd.DataFrame({\n",
    "    \"Метрика\": [\"MAE_train\", \"RMSE_train\", \"R2_train\", \"MAE_test\", \"RMSE_test\", \"R2_test\"],\n",
    "    \"Значение\": [\n",
    "        mean_absolute_error(y_train, pred_train),\n",
    "        mean_squared_error(y_train, pred_train) ** 0.5,\n",
    "        r2_score(y_train, pred_train),\n",
    "        mean_absolute_error(y_test, pred_test),\n",
    "        mean_squared_error(y_test, pred_test) ** 0.5,\n",
    "        r2_score(y_test, pred_test)\n",
    "    ]\n",
    "})\n",
    "display(metrics_final)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c268c783",
   "metadata": {},
   "source": [
    "## Визуализация новых переменных"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "7a971f2b",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 800x600 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "\n",
    "plt.figure(figsize=(8, 6))\n",
    "plt.scatter(X_best[\"Comp1\"], X_best[\"Comp2\"], alpha=0.25)\n",
    "plt.xlabel(\"Comp1\")\n",
    "plt.ylabel(\"Comp2\")\n",
    "plt.title(f\"Новые признаки после {best_method}\")\n",
    "plt.grid(alpha=0.3)\n",
    "plt.tight_layout()\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c6d7233d",
   "metadata": {},
   "source": [
    "## Итог"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4b9c4c66",
   "metadata": {},
   "source": [
    "В качестве целевой переменной использована saleEstimate_currentPrice.\n",
    "Для анализа оставлены 14 независимых числовых признаков.\n",
    "\n",
    "Снижение размерности\n",
    "\n",
    "Были протестированы методы:\n",
    "\n",
    "PCA\n",
    "\n",
    "Truncated SVD\n",
    "\n",
    "Factor Analysis\n",
    "\n",
    "Лучший результат показал Factor Analysis:\n",
    "\n",
    "Метод\tR²_test\n",
    "FactorAnalysis\t0.960\n",
    "PCA\t0.845\n",
    "TruncatedSVD\t0.845\n",
    "\n",
    "Поэтому лучшим методом снижения размерности признан Factor Analysis.\n",
    "\n",
    "Новые переменные\n",
    "\n",
    "После применения метода были сформированы два новых признака:\n",
    "\n",
    "Comp1\n",
    "\n",
    "Comp2\n",
    "\n",
    "Машинное обучение\n",
    "\n",
    "По новым признакам решена задача регрессии цены недвижимости.\n",
    "\n",
    "Качество модели:\n",
    "\n",
    "MAE_test ≈ 96 334\n",
    "\n",
    "RMSE_test ≈ 184 831\n",
    "\n",
    "R²_test ≈ 0.960\n",
    "\n",
    "Вывод\n",
    "\n",
    "Лучший метод снижения размерности — Factor Analysis.\n",
    "Регрессия по двум новым переменным сохраняет высокую точность (R² ≈ 0.96), что показывает, что большая часть информации о цене недвижимости успешно сохраняется после снижения размерности. "
   ]
  }
 ],
 "metadata": {
  "language_info": {
   "name": "python"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}