RealEstatePricePredictionMoscow03.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "6c895a6a",
   "metadata": {},
   "outputs": [],
   "source": [
    "# импорт используемых библиотек\n",
    "\n",
    "import numpy as np\n",
    "import pandas as pd\n",
    "import random\n",
    "\n",
    "from sklearn.model_selection import train_test_split, cross_val_score\n",
    "from sklearn.preprocessing import StandardScaler, RobustScaler\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.metrics import r2_score as r2\n",
    "from sklearn.model_selection import KFold, GridSearchCV\n",
    "\n",
    "from datetime import datetime\n",
    "\n",
    "import matplotlib\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n",
    "\n",
    "# отключаем предупреждения\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "\n",
    "# устанавливаемый единый размер шрифта\n",
    "matplotlib.rcParams.update({'font.size': 14})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "e639b2ff",
   "metadata": {},
   "outputs": [],
   "source": [
    "# функция для визуальной валидации результатов предсказания\n",
    "\n",
    "def evaluate_preds(train_true_values, train_pred_values, test_true_values, test_pred_values):\n",
    "    print(\"Train R2:\\t\" + str(round(r2(train_true_values, train_pred_values), 3)))\n",
    "    print(\"Test R2:\\t\" + str(round(r2(test_true_values, test_pred_values), 3)))\n",
    "    \n",
    "    plt.figure(figsize=(18,10))\n",
    "    \n",
    "    plt.subplot(121)\n",
    "    sns.scatterplot(x=train_pred_values, y=train_true_values)\n",
    "    plt.xlabel('Predicted values')\n",
    "    plt.ylabel('True values')\n",
    "    plt.title('Train sample prediction')\n",
    "    \n",
    "    plt.subplot(122)\n",
    "    sns.scatterplot(x=test_pred_values, y=test_true_values)\n",
    "    plt.xlabel('Predicted values')\n",
    "    plt.ylabel('True values')\n",
    "    plt.title('Test sample prediction')\n",
    "\n",
    "    plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "45009ed5",
   "metadata": {},
   "outputs": [],
   "source": [
    "TRAIN_DATASET_PATH = 'C:/ARTEM/GeekBrains/Python4DS/RealEstatePricePredictionMoscow/train.csv'\n",
    "TEST_DATASET_PATH = 'C:/ARTEM/GeekBrains/Python4DS/RealEstatePricePredictionMoscow/test.csv'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "14b0f3fc",
   "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>Id</th>\n",
       "      <th>DistrictId</th>\n",
       "      <th>Rooms</th>\n",
       "      <th>Square</th>\n",
       "      <th>LifeSquare</th>\n",
       "      <th>KitchenSquare</th>\n",
       "      <th>Floor</th>\n",
       "      <th>HouseFloor</th>\n",
       "      <th>HouseYear</th>\n",
       "      <th>Ecology_1</th>\n",
       "      <th>Ecology_2</th>\n",
       "      <th>Ecology_3</th>\n",
       "      <th>Social_1</th>\n",
       "      <th>Social_2</th>\n",
       "      <th>Social_3</th>\n",
       "      <th>Healthcare_1</th>\n",
       "      <th>Helthcare_2</th>\n",
       "      <th>Shops_1</th>\n",
       "      <th>Shops_2</th>\n",
       "      <th>Price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>9995</th>\n",
       "      <td>1260</td>\n",
       "      <td>61</td>\n",
       "      <td>2.0</td>\n",
       "      <td>49.090728</td>\n",
       "      <td>33.272626</td>\n",
       "      <td>6.0</td>\n",
       "      <td>3</td>\n",
       "      <td>12.0</td>\n",
       "      <td>1981</td>\n",
       "      <td>0.300323</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>52</td>\n",
       "      <td>10311</td>\n",
       "      <td>6</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>B</td>\n",
       "      <td>119367.455796</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9996</th>\n",
       "      <td>16265</td>\n",
       "      <td>27</td>\n",
       "      <td>2.0</td>\n",
       "      <td>64.307684</td>\n",
       "      <td>37.038420</td>\n",
       "      <td>9.0</td>\n",
       "      <td>13</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1977</td>\n",
       "      <td>0.072158</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>2</td>\n",
       "      <td>629</td>\n",
       "      <td>1</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>A</td>\n",
       "      <td>199715.148807</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9997</th>\n",
       "      <td>2795</td>\n",
       "      <td>178</td>\n",
       "      <td>1.0</td>\n",
       "      <td>29.648057</td>\n",
       "      <td>16.555363</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1958</td>\n",
       "      <td>0.460556</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>20</td>\n",
       "      <td>4386</td>\n",
       "      <td>14</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>5</td>\n",
       "      <td>B</td>\n",
       "      <td>165953.912580</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9998</th>\n",
       "      <td>14561</td>\n",
       "      <td>21</td>\n",
       "      <td>1.0</td>\n",
       "      <td>32.330292</td>\n",
       "      <td>22.326870</td>\n",
       "      <td>5.0</td>\n",
       "      <td>3</td>\n",
       "      <td>9.0</td>\n",
       "      <td>1969</td>\n",
       "      <td>0.194489</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>47</td>\n",
       "      <td>8004</td>\n",
       "      <td>3</td>\n",
       "      <td>125.0</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>B</td>\n",
       "      <td>171842.411855</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9999</th>\n",
       "      <td>7202</td>\n",
       "      <td>94</td>\n",
       "      <td>1.0</td>\n",
       "      <td>35.815476</td>\n",
       "      <td>22.301367</td>\n",
       "      <td>6.0</td>\n",
       "      <td>9</td>\n",
       "      <td>9.0</td>\n",
       "      <td>1975</td>\n",
       "      <td>0.127376</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>43</td>\n",
       "      <td>8429</td>\n",
       "      <td>3</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>9</td>\n",
       "      <td>B</td>\n",
       "      <td>177685.627486</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Id  DistrictId  Rooms     Square  LifeSquare  KitchenSquare  Floor  \\\n",
       "9995   1260          61    2.0  49.090728   33.272626            6.0      3   \n",
       "9996  16265          27    2.0  64.307684   37.038420            9.0     13   \n",
       "9997   2795         178    1.0  29.648057   16.555363            5.0      3   \n",
       "9998  14561          21    1.0  32.330292   22.326870            5.0      3   \n",
       "9999   7202          94    1.0  35.815476   22.301367            6.0      9   \n",
       "\n",
       "      HouseFloor  HouseYear  Ecology_1 Ecology_2 Ecology_3  Social_1  \\\n",
       "9995        12.0       1981   0.300323         B         B        52   \n",
       "9996         0.0       1977   0.072158         B         B         2   \n",
       "9997         5.0       1958   0.460556         B         B        20   \n",
       "9998         9.0       1969   0.194489         B         B        47   \n",
       "9999         9.0       1975   0.127376         B         B        43   \n",
       "\n",
       "      Social_2  Social_3  Healthcare_1  Helthcare_2  Shops_1 Shops_2  \\\n",
       "9995     10311         6           NaN            1        9       B   \n",
       "9996       629         1           NaN            0        0       A   \n",
       "9997      4386        14           NaN            1        5       B   \n",
       "9998      8004         3         125.0            3        5       B   \n",
       "9999      8429         3           NaN            3        9       B   \n",
       "\n",
       "              Price  \n",
       "9995  119367.455796  \n",
       "9996  199715.148807  \n",
       "9997  165953.912580  \n",
       "9998  171842.411855  \n",
       "9999  177685.627486  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# загружаем обучающую выборку, проверяем данные\n",
    "train_df = pd.read_csv(TRAIN_DATASET_PATH)\n",
    "train_df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "18438e1c",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Id                 int64\n",
       "DistrictId         int64\n",
       "Rooms            float64\n",
       "Square           float64\n",
       "LifeSquare       float64\n",
       "KitchenSquare    float64\n",
       "Floor              int64\n",
       "HouseFloor       float64\n",
       "HouseYear          int64\n",
       "Ecology_1        float64\n",
       "Ecology_2         object\n",
       "Ecology_3         object\n",
       "Social_1           int64\n",
       "Social_2           int64\n",
       "Social_3           int64\n",
       "Healthcare_1     float64\n",
       "Helthcare_2        int64\n",
       "Shops_1            int64\n",
       "Shops_2           object\n",
       "Price            float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# смотрим типы данных, цель уменьшить размер датасета\n",
    "train_df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "e790eca4",
   "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>Id</th>\n",
       "      <th>DistrictId</th>\n",
       "      <th>Rooms</th>\n",
       "      <th>Square</th>\n",
       "      <th>LifeSquare</th>\n",
       "      <th>KitchenSquare</th>\n",
       "      <th>Floor</th>\n",
       "      <th>HouseFloor</th>\n",
       "      <th>HouseYear</th>\n",
       "      <th>Ecology_1</th>\n",
       "      <th>Ecology_2</th>\n",
       "      <th>Ecology_3</th>\n",
       "      <th>Social_1</th>\n",
       "      <th>Social_2</th>\n",
       "      <th>Social_3</th>\n",
       "      <th>Healthcare_1</th>\n",
       "      <th>Helthcare_2</th>\n",
       "      <th>Shops_1</th>\n",
       "      <th>Shops_2</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4995</th>\n",
       "      <td>10379</td>\n",
       "      <td>29</td>\n",
       "      <td>2.0</td>\n",
       "      <td>43.177521</td>\n",
       "      <td>30.339945</td>\n",
       "      <td>5.0</td>\n",
       "      <td>6</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1962</td>\n",
       "      <td>0.069660</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>31</td>\n",
       "      <td>6119</td>\n",
       "      <td>4</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>B</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4996</th>\n",
       "      <td>16138</td>\n",
       "      <td>38</td>\n",
       "      <td>3.0</td>\n",
       "      <td>93.698122</td>\n",
       "      <td>94.521465</td>\n",
       "      <td>10.0</td>\n",
       "      <td>21</td>\n",
       "      <td>27.0</td>\n",
       "      <td>2018</td>\n",
       "      <td>0.060753</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>15</td>\n",
       "      <td>2787</td>\n",
       "      <td>2</td>\n",
       "      <td>520.0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>B</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4997</th>\n",
       "      <td>3912</td>\n",
       "      <td>101</td>\n",
       "      <td>1.0</td>\n",
       "      <td>33.656723</td>\n",
       "      <td>19.003259</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1966</td>\n",
       "      <td>0.038693</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>28</td>\n",
       "      <td>6533</td>\n",
       "      <td>1</td>\n",
       "      <td>1015.0</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>B</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4998</th>\n",
       "      <td>5722</td>\n",
       "      <td>10</td>\n",
       "      <td>1.0</td>\n",
       "      <td>38.635155</td>\n",
       "      <td>20.976257</td>\n",
       "      <td>9.0</td>\n",
       "      <td>8</td>\n",
       "      <td>14.0</td>\n",
       "      <td>1970</td>\n",
       "      <td>0.089040</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>33</td>\n",
       "      <td>7976</td>\n",
       "      <td>5</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>B</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4999</th>\n",
       "      <td>11004</td>\n",
       "      <td>21</td>\n",
       "      <td>2.0</td>\n",
       "      <td>67.122742</td>\n",
       "      <td>33.944344</td>\n",
       "      <td>13.0</td>\n",
       "      <td>9</td>\n",
       "      <td>17.0</td>\n",
       "      <td>2009</td>\n",
       "      <td>0.194489</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>47</td>\n",
       "      <td>8004</td>\n",
       "      <td>3</td>\n",
       "      <td>125.0</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>B</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Id  DistrictId  Rooms     Square  LifeSquare  KitchenSquare  Floor  \\\n",
       "4995  10379          29    2.0  43.177521   30.339945            5.0      6   \n",
       "4996  16138          38    3.0  93.698122   94.521465           10.0     21   \n",
       "4997   3912         101    1.0  33.656723   19.003259            5.0      2   \n",
       "4998   5722          10    1.0  38.635155   20.976257            9.0      8   \n",
       "4999  11004          21    2.0  67.122742   33.944344           13.0      9   \n",
       "\n",
       "      HouseFloor  HouseYear  Ecology_1 Ecology_2 Ecology_3  Social_1  \\\n",
       "4995         5.0       1962   0.069660         B         B        31   \n",
       "4996        27.0       2018   0.060753         B         B        15   \n",
       "4997         5.0       1966   0.038693         B         B        28   \n",
       "4998        14.0       1970   0.089040         B         B        33   \n",
       "4999        17.0       2009   0.194489         B         B        47   \n",
       "\n",
       "      Social_2  Social_3  Healthcare_1  Helthcare_2  Shops_1 Shops_2  \n",
       "4995      6119         4           NaN            1        2       B  \n",
       "4996      2787         2         520.0            0        7       B  \n",
       "4997      6533         1        1015.0            2        5       B  \n",
       "4998      7976         5           NaN            0       11       B  \n",
       "4999      8004         3         125.0            3        5       B  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# загружаем проверочную выборку, проверяем данные\n",
    "\n",
    "test_df = pd.read_csv(TEST_DATASET_PATH)\n",
    "test_df.tail()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "ae193efd",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Строк в трейне: 10000\n",
      "Строк в тесте 5000\n"
     ]
    }
   ],
   "source": [
    "# размеры датасетов\n",
    "\n",
    "print('Строк в трейне:', train_df.shape[0])\n",
    "print('Строк в тесте', test_df.shape[0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "ec2ebfcf",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# проверяем размерности датасетов - обучающий нимеет на столбец больше проверочного\n",
    "train_df.shape[1] - 1 == test_df.shape[1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "17395fc9",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Id                object\n",
       "DistrictId        object\n",
       "Rooms            float64\n",
       "Square           float64\n",
       "LifeSquare       float64\n",
       "KitchenSquare    float64\n",
       "Floor              int16\n",
       "HouseFloor         int16\n",
       "HouseYear          int16\n",
       "Ecology_1        float64\n",
       "Ecology_2         object\n",
       "Ecology_3         object\n",
       "Social_1           int64\n",
       "Social_2           int64\n",
       "Social_3           int64\n",
       "Healthcare_1     float64\n",
       "Helthcare_2        int64\n",
       "Shops_1            int64\n",
       "Shops_2           object\n",
       "Price            float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# приведение типов к строкам, чтобы модель не применяла <> для этих данных\n",
    "train_df['Id'] = train_df['Id'].astype(str)\n",
    "train_df['DistrictId'] = train_df['DistrictId'].astype(str)\n",
    "\n",
    "# уменьшаем размерности этажей,этажностей, возраста\n",
    "train_df['Floor'] = train_df['Floor'].astype(np.int16)\n",
    "train_df['HouseFloor'] = train_df['HouseFloor'].astype(np.int16)\n",
    "train_df['HouseYear'] = train_df['HouseYear'].astype(np.int16)\n",
    "\n",
    "\n",
    "# проверяем преобразование\n",
    "train_df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "c6491ecb",
   "metadata": {},
   "outputs": [],
   "source": [
    "### Исследуем данные"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "5ff7bb43",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# гистограма распределения целевых значений\n",
    "plt.figure(figsize = (16, 8))\n",
    "\n",
    "train_df['Price'].hist(bins=30)\n",
    "plt.ylabel('Count')\n",
    "plt.xlabel('Price')\n",
    "\n",
    "plt.title('Target distribution')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "07ea1861",
   "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>Rooms</th>\n",
       "      <th>Square</th>\n",
       "      <th>LifeSquare</th>\n",
       "      <th>KitchenSquare</th>\n",
       "      <th>Floor</th>\n",
       "      <th>HouseFloor</th>\n",
       "      <th>HouseYear</th>\n",
       "      <th>Ecology_1</th>\n",
       "      <th>Social_1</th>\n",
       "      <th>Social_2</th>\n",
       "      <th>Social_3</th>\n",
       "      <th>Healthcare_1</th>\n",
       "      <th>Helthcare_2</th>\n",
       "      <th>Shops_1</th>\n",
       "      <th>Price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>7887.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>5202.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.890500</td>\n",
       "      <td>56.315775</td>\n",
       "      <td>37.199645</td>\n",
       "      <td>6.273300</td>\n",
       "      <td>8.526700</td>\n",
       "      <td>12.609400</td>\n",
       "      <td>1984.764700</td>\n",
       "      <td>0.118858</td>\n",
       "      <td>24.687000</td>\n",
       "      <td>5352.157400</td>\n",
       "      <td>8.039200</td>\n",
       "      <td>1142.904460</td>\n",
       "      <td>1.319500</td>\n",
       "      <td>4.231300</td>\n",
       "      <td>214138.857399</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.839512</td>\n",
       "      <td>21.058732</td>\n",
       "      <td>86.241209</td>\n",
       "      <td>28.560917</td>\n",
       "      <td>5.241148</td>\n",
       "      <td>6.775974</td>\n",
       "      <td>53.112756</td>\n",
       "      <td>0.119025</td>\n",
       "      <td>17.532614</td>\n",
       "      <td>4006.799803</td>\n",
       "      <td>23.831875</td>\n",
       "      <td>1021.517264</td>\n",
       "      <td>1.493601</td>\n",
       "      <td>4.806341</td>\n",
       "      <td>92872.293865</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.136859</td>\n",
       "      <td>0.370619</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-2005.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>168.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>59174.778028</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>41.774881</td>\n",
       "      <td>22.769832</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>1974.000000</td>\n",
       "      <td>0.017647</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>1564.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>350.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>153872.633942</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>2.000000</td>\n",
       "      <td>52.513310</td>\n",
       "      <td>32.781260</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>1977.000000</td>\n",
       "      <td>0.075424</td>\n",
       "      <td>25.000000</td>\n",
       "      <td>5285.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>900.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>192269.644879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>2.000000</td>\n",
       "      <td>65.900625</td>\n",
       "      <td>45.128803</td>\n",
       "      <td>9.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>17.000000</td>\n",
       "      <td>2001.000000</td>\n",
       "      <td>0.195781</td>\n",
       "      <td>36.000000</td>\n",
       "      <td>7227.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>1548.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>249135.462171</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>19.000000</td>\n",
       "      <td>641.065193</td>\n",
       "      <td>7480.592129</td>\n",
       "      <td>2014.000000</td>\n",
       "      <td>42.000000</td>\n",
       "      <td>117.000000</td>\n",
       "      <td>4968.000000</td>\n",
       "      <td>0.521867</td>\n",
       "      <td>74.000000</td>\n",
       "      <td>19083.000000</td>\n",
       "      <td>141.000000</td>\n",
       "      <td>4849.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>23.000000</td>\n",
       "      <td>633233.466570</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "              Rooms        Square   LifeSquare  KitchenSquare         Floor  \\\n",
       "count  10000.000000  10000.000000  7887.000000   10000.000000  10000.000000   \n",
       "mean       1.890500     56.315775    37.199645       6.273300      8.526700   \n",
       "std        0.839512     21.058732    86.241209      28.560917      5.241148   \n",
       "min        0.000000      1.136859     0.370619       0.000000      1.000000   \n",
       "25%        1.000000     41.774881    22.769832       1.000000      4.000000   \n",
       "50%        2.000000     52.513310    32.781260       6.000000      7.000000   \n",
       "75%        2.000000     65.900625    45.128803       9.000000     12.000000   \n",
       "max       19.000000    641.065193  7480.592129    2014.000000     42.000000   \n",
       "\n",
       "         HouseFloor     HouseYear     Ecology_1      Social_1      Social_2  \\\n",
       "count  10000.000000  10000.000000  10000.000000  10000.000000  10000.000000   \n",
       "mean      12.609400   1984.764700      0.118858     24.687000   5352.157400   \n",
       "std        6.775974     53.112756      0.119025     17.532614   4006.799803   \n",
       "min        0.000000  -2005.000000      0.000000      0.000000    168.000000   \n",
       "25%        9.000000   1974.000000      0.017647      6.000000   1564.000000   \n",
       "50%       13.000000   1977.000000      0.075424     25.000000   5285.000000   \n",
       "75%       17.000000   2001.000000      0.195781     36.000000   7227.000000   \n",
       "max      117.000000   4968.000000      0.521867     74.000000  19083.000000   \n",
       "\n",
       "           Social_3  Healthcare_1   Helthcare_2       Shops_1          Price  \n",
       "count  10000.000000   5202.000000  10000.000000  10000.000000   10000.000000  \n",
       "mean       8.039200   1142.904460      1.319500      4.231300  214138.857399  \n",
       "std       23.831875   1021.517264      1.493601      4.806341   92872.293865  \n",
       "min        0.000000      0.000000      0.000000      0.000000   59174.778028  \n",
       "25%        0.000000    350.000000      0.000000      1.000000  153872.633942  \n",
       "50%        2.000000    900.000000      1.000000      3.000000  192269.644879  \n",
       "75%        5.000000   1548.000000      2.000000      6.000000  249135.462171  \n",
       "max      141.000000   4849.000000      6.000000     23.000000  633233.466570  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# смотрим на параметры для данных выборки\n",
    "train_df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "983e0cfd",
   "metadata": {},
   "outputs": [],
   "source": [
    "# количество комнат \n",
    "# 0 - допустимое значение для студии\n",
    "# площадь"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "8923291f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Id', 'DistrictId', 'Ecology_2', 'Ecology_3', 'Shops_2']"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df.select_dtypes(include='object').columns.tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "30f9ec22",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "27     851\n",
       "1      652\n",
       "23     565\n",
       "6      511\n",
       "9      294\n",
       "      ... \n",
       "199      1\n",
       "117      1\n",
       "207      1\n",
       "209      1\n",
       "174      1\n",
       "Name: DistrictId, Length: 205, dtype: int64"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df['DistrictId'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "81c3d0f4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "B    9903\n",
       "A      97\n",
       "Name: Ecology_2, dtype: int64"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df['Ecology_2'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "80b59c3a",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "B    9725\n",
       "A     275\n",
       "Name: Ecology_3, dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df['Ecology_3'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "bb41a69d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "B    9175\n",
       "A     825\n",
       "Name: Shops_2, dtype: int64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df['Shops_2'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "7761c352",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Обработка выбросов\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "20aabf1e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2.0     3880\n",
       "1.0     3705\n",
       "3.0     2235\n",
       "4.0      150\n",
       "5.0       18\n",
       "0.0        8\n",
       "10.0       2\n",
       "19.0       1\n",
       "6.0        1\n",
       "Name: Rooms, dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df['Rooms'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "5acd823b",
   "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>Id</th>\n",
       "      <th>DistrictId</th>\n",
       "      <th>Rooms</th>\n",
       "      <th>Square</th>\n",
       "      <th>LifeSquare</th>\n",
       "      <th>KitchenSquare</th>\n",
       "      <th>Floor</th>\n",
       "      <th>HouseFloor</th>\n",
       "      <th>HouseYear</th>\n",
       "      <th>Ecology_1</th>\n",
       "      <th>Ecology_2</th>\n",
       "      <th>Ecology_3</th>\n",
       "      <th>Social_1</th>\n",
       "      <th>Social_2</th>\n",
       "      <th>Social_3</th>\n",
       "      <th>Healthcare_1</th>\n",
       "      <th>Helthcare_2</th>\n",
       "      <th>Shops_1</th>\n",
       "      <th>Shops_2</th>\n",
       "      <th>Price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11809</td>\n",
       "      <td>27</td>\n",
       "      <td>3.0</td>\n",
       "      <td>115.027311</td>\n",
       "      <td>NaN</td>\n",
       "      <td>10.0</td>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>2014</td>\n",
       "      <td>0.075424</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>11</td>\n",
       "      <td>3097</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>B</td>\n",
       "      <td>305018.871089</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3013</td>\n",
       "      <td>22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>39.832524</td>\n",
       "      <td>23.169223</td>\n",
       "      <td>8.0</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "      <td>1966</td>\n",
       "      <td>0.118537</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>30</td>\n",
       "      <td>6207</td>\n",
       "      <td>1</td>\n",
       "      <td>1183.0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>B</td>\n",
       "      <td>177734.553407</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8215</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>78.342215</td>\n",
       "      <td>47.671972</td>\n",
       "      <td>10.0</td>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>1988</td>\n",
       "      <td>0.025609</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>33</td>\n",
       "      <td>5261</td>\n",
       "      <td>0</td>\n",
       "      <td>240.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>B</td>\n",
       "      <td>282078.720850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2352</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>40.409907</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.0</td>\n",
       "      <td>10</td>\n",
       "      <td>22</td>\n",
       "      <td>1977</td>\n",
       "      <td>0.007122</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>1</td>\n",
       "      <td>264</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>B</td>\n",
       "      <td>168106.007630</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>13866</td>\n",
       "      <td>94</td>\n",
       "      <td>2.0</td>\n",
       "      <td>64.285067</td>\n",
       "      <td>38.562517</td>\n",
       "      <td>9.0</td>\n",
       "      <td>16</td>\n",
       "      <td>16</td>\n",
       "      <td>1972</td>\n",
       "      <td>0.282798</td>\n",
       "      <td>B</td>\n",
       "      <td>B</td>\n",
       "      <td>33</td>\n",
       "      <td>8667</td>\n",
       "      <td>2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>B</td>\n",
       "      <td>343995.102962</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      Id DistrictId  Rooms      Square  LifeSquare  KitchenSquare  Floor  \\\n",
       "0  11809         27    3.0  115.027311         NaN           10.0      4   \n",
       "1   3013         22    1.0   39.832524   23.169223            8.0      7   \n",
       "2   8215          1    3.0   78.342215   47.671972           10.0      2   \n",
       "3   2352          1    1.0   40.409907         NaN            1.0     10   \n",
       "4  13866         94    2.0   64.285067   38.562517            9.0     16   \n",
       "\n",
       "   HouseFloor  HouseYear  Ecology_1 Ecology_2 Ecology_3  Social_1  Social_2  \\\n",
       "0          10       2014   0.075424         B         B        11      3097   \n",
       "1           8       1966   0.118537         B         B        30      6207   \n",
       "2          17       1988   0.025609         B         B        33      5261   \n",
       "3          22       1977   0.007122         B         B         1       264   \n",
       "4          16       1972   0.282798         B         B        33      8667   \n",
       "\n",
       "   Social_3  Healthcare_1  Helthcare_2  Shops_1 Shops_2          Price  \n",
       "0         0           NaN            0        0       B  305018.871089  \n",
       "1         1        1183.0            1        0       B  177734.553407  \n",
       "2         0         240.0            3        1       B  282078.720850  \n",
       "3         0           NaN            0        1       B  168106.007630  \n",
       "4         2           NaN            0        6       B  343995.102962  "
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# пример, где считаем выбросами количество комнат равное 0 или >=6: создаем столбец Rooms_outlier, \n",
    "# где указываем недоверие к числу комнат\n",
    "\n",
    "#train_df['Rooms_outlier'] = 0\n",
    "#train_df.loc[(train_df['Rooms'] == 0) | (train_df['Rooms'] >= 6), 'Rooms_outlier'] = 1\n",
    "#train_df.loc[train_df['Rooms'] == 0, 'Rooms'] = 1\n",
    "#train_df.loc[train_df['Rooms'] >= 6, 'Rooms'] = train_df['Rooms'].median()\n",
    "\n",
    "train_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "8c897557",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2.0     3880\n",
       "1.0     3705\n",
       "3.0     2235\n",
       "4.0      150\n",
       "5.0       18\n",
       "0.0        8\n",
       "10.0       2\n",
       "19.0       1\n",
       "6.0        1\n",
       "Name: Rooms, dtype: int64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df['Rooms'].value_counts()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "070297c7",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1.0       2460\n",
       "8.0       1306\n",
       "5.0       1169\n",
       "10.0      1075\n",
       "6.0       1038\n",
       "9.0        843\n",
       "0.0        697\n",
       "7.0        609\n",
       "12.0       249\n",
       "11.0       233\n",
       "13.0        67\n",
       "14.0        51\n",
       "4.0         39\n",
       "15.0        31\n",
       "3.0         22\n",
       "16.0        16\n",
       "20.0        14\n",
       "17.0        12\n",
       "19.0        11\n",
       "18.0         6\n",
       "2.0          4\n",
       "22.0         3\n",
       "30.0         2\n",
       "43.0         2\n",
       "41.0         2\n",
       "112.0        2\n",
       "25.0         2\n",
       "51.0         2\n",
       "37.0         2\n",
       "58.0         2\n",
       "32.0         2\n",
       "21.0         1\n",
       "73.0         1\n",
       "75.0         1\n",
       "36.0         1\n",
       "27.0         1\n",
       "63.0         1\n",
       "1970.0       1\n",
       "54.0         1\n",
       "53.0         1\n",
       "60.0         1\n",
       "26.0         1\n",
       "66.0         1\n",
       "39.0         1\n",
       "29.0         1\n",
       "78.0         1\n",
       "31.0         1\n",
       "84.0         1\n",
       "48.0         1\n",
       "96.0         1\n",
       "42.0         1\n",
       "40.0         1\n",
       "23.0         1\n",
       "72.0         1\n",
       "35.0         1\n",
       "62.0         1\n",
       "123.0        1\n",
       "2014.0       1\n",
       "Name: KitchenSquare, dtype: int64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df['KitchenSquare'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "d0c61fac",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(13.0, 0.0)"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df['KitchenSquare'].quantile(.975), train_df['KitchenSquare'].quantile(.025)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "e64fb693",
   "metadata": {},
   "outputs": [],
   "source": [
    "# обрабатываем выбросы в площадях кухни\n",
    "condition = (train_df['KitchenSquare'].isna()) \\\n",
    "             | (train_df['KitchenSquare'] > train_df['KitchenSquare'].quantile(.975))\n",
    "        \n",
    "train_df.loc[condition, 'KitchenSquare'] = train_df['KitchenSquare'].median()\n",
    "\n",
    "train_df.loc[train_df['KitchenSquare'] < 3, 'KitchenSquare'] = 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "4e632236",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3.0     3183\n",
       "8.0     1306\n",
       "6.0     1227\n",
       "5.0     1169\n",
       "10.0    1075\n",
       "9.0      843\n",
       "7.0      609\n",
       "12.0     249\n",
       "11.0     233\n",
       "13.0      67\n",
       "4.0       39\n",
       "Name: KitchenSquare, dtype: int64"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df['KitchenSquare'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "38abe743",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([  0,   1,   2,   3,   4,   5,   6,   7,   8,   9,  10,  11,  12,\n",
       "        13,  14,  15,  16,  17,  18,  19,  20,  21,  22,  23,  24,  25,\n",
       "        26,  27,  28,  29,  30,  31,  32,  36,  37,  38,  39,  40,  44,\n",
       "        45,  47,  48,  99, 117], dtype=int16)"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# исследуем этажность зданий\n",
    "\n",
    "train_df['HouseFloor'].sort_values().unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "03135765",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([ 1,  2,  3,  4,  5,  6,  7,  8,  9, 10, 11, 12, 13, 14, 15, 16, 17,\n",
       "       18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 31, 32, 33, 37, 42],\n",
       "      dtype=int16)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# исследуем этаж\n",
    "\n",
    "train_df['Floor'].sort_values().unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "d2b2bb71",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1825"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# находим случаи, когда этаж больше этажности здания\n",
    "(train_df['Floor'] > train_df['HouseFloor']).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "b5f769f4",
   "metadata": {},
   "outputs": [],
   "source": [
    "# добавляем признак достоверности этажа\n",
    "train_df['HouseFloor_outlier'] = 0\n",
    "train_df.loc[train_df['HouseFloor'] == 0, 'HouseFloor_outlier'] = 1\n",
    "train_df.loc[train_df['Floor'] > train_df['HouseFloor'], 'HouseFloor_outlier'] = 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "863234a8",
   "metadata": {},
   "outputs": [],
   "source": [
    "# некорректную этажность заменяем медианой\n",
    "train_df.loc[train_df['HouseFloor'] == 0, 'HouseFloor'] = train_df['HouseFloor'].median()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "35b81479",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Int64Index([  17,   19,   21,   25,   26,   33,   34,   37,   39,   44,\n",
       "            ...\n",
       "            9913, 9921, 9930, 9931, 9938, 9953, 9960, 9968, 9970, 9975],\n",
       "           dtype='int64', length=1610)"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# смотрим на id проблемных записей по этажам\n",
    "floor_outliers = train_df.loc[train_df['Floor'] > train_df['HouseFloor']].index\n",
    "floor_outliers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "e779720c",
   "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>Price</th>\n",
       "      <th>Square</th>\n",
       "      <th>Meter_price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "      <td>10000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>214138.857399</td>\n",
       "      <td>56.315775</td>\n",
       "      <td>3988.383777</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>92872.293865</td>\n",
       "      <td>21.058732</td>\n",
       "      <td>3755.437224</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>59174.778028</td>\n",
       "      <td>1.136859</td>\n",
       "      <td>208.293451</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>153872.633942</td>\n",
       "      <td>41.774881</td>\n",
       "      <td>2918.525328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>192269.644879</td>\n",
       "      <td>52.513310</td>\n",
       "      <td>3910.052336</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>249135.462171</td>\n",
       "      <td>65.900625</td>\n",
       "      <td>4692.375993</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>633233.466570</td>\n",
       "      <td>641.065193</td>\n",
       "      <td>230463.537966</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               Price        Square    Meter_price\n",
       "count   10000.000000  10000.000000   10000.000000\n",
       "mean   214138.857399     56.315775    3988.383777\n",
       "std     92872.293865     21.058732    3755.437224\n",
       "min     59174.778028      1.136859     208.293451\n",
       "25%    153872.633942     41.774881    2918.525328\n",
       "50%    192269.644879     52.513310    3910.052336\n",
       "75%    249135.462171     65.900625    4692.375993\n",
       "max    633233.466570    641.065193  230463.537966"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Добавляем цену квадратного метра AV\n",
    "train_df['Meter_price'] = 0\n",
    "train_df.loc[train_df['Square'] > 0, 'Meter_price'] = train_df['Price'] / train_df['Square']\n",
    "#train_df['Meter_price'].sort_values(ascending=False)\n",
    "train_df[['Price', 'Square', 'Meter_price']].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "9f6e136e",
   "metadata": {},
   "outputs": [],
   "source": [
    "# подстановка для значений этажа\n",
    "train_df.loc[floor_outliers, 'Floor'] = train_df.loc[floor_outliers, 'HouseFloor']\\\n",
    "                                                .apply(lambda x: random.randint(1, x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "37b4218f",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "(train_df['Floor'] > train_df['HouseFloor']).sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "a83f6692",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "8828    4968\n",
       "6073    2020\n",
       "5641    2020\n",
       "2350    2020\n",
       "9091    2020\n",
       "        ... \n",
       "6250    1914\n",
       "2066    1912\n",
       "2795    1912\n",
       "6794    1910\n",
       "9163   -2005\n",
       "Name: HouseYear, Length: 10000, dtype: int16"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# выводим года постройки\n",
    "train_df['HouseYear'].sort_values(ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "b5f4456f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# если год постройки еще не наступал, то меняем на текущий, можно менять на медиану\n",
    "train_df.loc[train_df['HouseYear'] > 2021, 'HouseYear'] = 2021\n",
    "# если дом слишком древний, до 1910 - это первый разумный минимум. Заменяем медианой\n",
    "train_df.loc[train_df['HouseYear'] < 1910, 'HouseYear'] = train_df['HouseYear'].median()\n",
    "# год можно также предсказывать по этажности, по району и т.д.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "9c753033",
   "metadata": {},
   "outputs": [],
   "source": [
    "# обработка пропусков"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "f310a637",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Id                       0\n",
       "DistrictId               0\n",
       "Rooms                    0\n",
       "Square                   0\n",
       "LifeSquare            2113\n",
       "KitchenSquare            0\n",
       "Floor                    0\n",
       "HouseFloor               0\n",
       "HouseYear                0\n",
       "Ecology_1                0\n",
       "Ecology_2                0\n",
       "Ecology_3                0\n",
       "Social_1                 0\n",
       "Social_2                 0\n",
       "Social_3                 0\n",
       "Healthcare_1          4798\n",
       "Helthcare_2              0\n",
       "Shops_1                  0\n",
       "Shops_2                  0\n",
       "Price                    0\n",
       "HouseFloor_outlier       0\n",
       "Meter_price              0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "d914ad63",
   "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>Square</th>\n",
       "      <th>LifeSquare</th>\n",
       "      <th>KitchenSquare</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>115.027311</td>\n",
       "      <td>NaN</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>39.832524</td>\n",
       "      <td>23.169223</td>\n",
       "      <td>8.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>78.342215</td>\n",
       "      <td>47.671972</td>\n",
       "      <td>10.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>40.409907</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>64.285067</td>\n",
       "      <td>38.562517</td>\n",
       "      <td>9.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>62.528465</td>\n",
       "      <td>47.103833</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>33.938750</td>\n",
       "      <td>21.772930</td>\n",
       "      <td>9.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>50.391814</td>\n",
       "      <td>32.893256</td>\n",
       "      <td>8.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>46.887892</td>\n",
       "      <td>44.628132</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>79.867477</td>\n",
       "      <td>75.085125</td>\n",
       "      <td>3.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       Square  LifeSquare  KitchenSquare\n",
       "0  115.027311         NaN           10.0\n",
       "1   39.832524   23.169223            8.0\n",
       "2   78.342215   47.671972           10.0\n",
       "3   40.409907         NaN            3.0\n",
       "4   64.285067   38.562517            9.0\n",
       "5   62.528465   47.103833            6.0\n",
       "6   33.938750   21.772930            9.0\n",
       "7   50.391814   32.893256            8.0\n",
       "8   46.887892   44.628132            3.0\n",
       "9   79.867477   75.085125            3.0"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df[['Square', 'LifeSquare', 'KitchenSquare']].head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "b68c9268",
   "metadata": {},
   "outputs": [],
   "source": [
    "# если нет данных по жилой площади - добавляем признак и пытаемся расчитать\n",
    "train_df['LifeSquare_nan'] = train_df['LifeSquare'].isna() * 1\n",
    "\n",
    "condition = (train_df['LifeSquare'].isna()) \\\n",
    "             & (~train_df['Square'].isna()) \\\n",
    "             & (~train_df['KitchenSquare'].isna())\n",
    "        \n",
    "train_df.loc[condition, 'LifeSquare'] = train_df.loc[condition, 'Square'] \\\n",
    "                                            - train_df.loc[condition, 'KitchenSquare'] - 3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "e1c3b780",
   "metadata": {},
   "outputs": [],
   "source": [
    "#train_df[['Healthcare_1','Healthcare_2']].head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "383b8664",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df.drop('Healthcare_1', axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "a113fc25",
   "metadata": {},
   "outputs": [],
   "source": [
    "class DataPreprocessing:\n",
    "    \"\"\"Подготовка исходных данных\"\"\"\n",
    "\n",
    "    def __init__(self):\n",
    "        \"\"\"Параметры класса\"\"\"\n",
    "        self.medians = None\n",
    "        self.kitchen_square_quantile = None\n",
    "        \n",
    "    def fit(self, X):\n",
    "        \"\"\"Сохранение статистик\"\"\"       \n",
    "        # Расчет медиан\n",
    "        self.medians = X.median()\n",
    "        self.kitchen_square_quantile = X['KitchenSquare'].quantile(.975)\n",
    "    \n",
    "    def transform(self, X):\n",
    "        \"\"\"Трансформация данных\"\"\"\n",
    "\n",
    "        # Rooms\n",
    "        X['Rooms_outlier'] = 0\n",
    "        X.loc[(X['Rooms'] == 0) | (X['Rooms'] >= 6), 'Rooms_outlier'] = 1\n",
    "        \n",
    "        X.loc[X['Rooms'] == 0, 'Rooms'] = 1\n",
    "        X.loc[X['Rooms'] >= 6, 'Rooms'] = self.medians['Rooms']\n",
    "        \n",
    "        # KitchenSquare\n",
    "        condition = (X['KitchenSquare'].isna()) \\\n",
    "                    | (X['KitchenSquare'] > self.kitchen_square_quantile)\n",
    "        \n",
    "        X.loc[condition, 'KitchenSquare'] = self.medians['KitchenSquare']\n",
    "\n",
    "        X.loc[X['KitchenSquare'] < 3, 'KitchenSquare'] = 3\n",
    "        \n",
    "        # HouseFloor, Floor\n",
    "        X['HouseFloor_outlier'] = 0\n",
    "        X.loc[X['HouseFloor'] == 0, 'HouseFloor_outlier'] = 1\n",
    "        X.loc[X['Floor'] > X['HouseFloor'], 'HouseFloor_outlier'] = 1\n",
    "        \n",
    "        X.loc[X['HouseFloor'] == 0, 'HouseFloor'] = self.medians['HouseFloor']\n",
    "        \n",
    "        floor_outliers = X.loc[X['Floor'] > X['HouseFloor']].index\n",
    "        X.loc[floor_outliers, 'Floor'] = X.loc[floor_outliers, 'HouseFloor']\\\n",
    "                                            .apply(lambda x: random.randint(1, x))\n",
    "        \n",
    "        # HouseYear\n",
    "        current_year = datetime.now().year\n",
    "        \n",
    "        X['HouseYear_outlier'] = 0\n",
    "        \n",
    "        X.loc[X['HouseYear'] > current_year, 'HouseYear_outlier'] = 1        \n",
    "        X.loc[X['HouseYear'] > current_year, 'HouseYear'] = current_year\n",
    "        \n",
    "        # AV\n",
    "        #==========================================================================\n",
    "        X.loc[X['HouseYear'] < 1910, 'HouseYear_outlier'] = 1 \n",
    "        X.loc[X['HouseYear'] < 1910, 'HouseYear'] = self.medians['HouseYear']\n",
    "        #==========================================================================\n",
    "        X.loc[X['Square'].isna(), 'Square'] = self.medians['Square']    \n",
    "        #==========================================================================\n",
    "    \n",
    "        \n",
    "        # Healthcare_1\n",
    "        if 'Healthcare_1' in X.columns:\n",
    "            X.drop('Healthcare_1', axis=1, inplace=True)\n",
    "            \n",
    "        # LifeSquare\n",
    "        X['LifeSquare_nan'] = X['LifeSquare'].isna() * 1\n",
    "        condition = (X['LifeSquare'].isna()) & \\\n",
    "                      (~X['Square'].isna()) & \\\n",
    "                      (~X['KitchenSquare'].isna())\n",
    "        \n",
    "        X.loc[condition, 'LifeSquare'] = X.loc[condition, 'Square'] - X.loc[condition, 'KitchenSquare'] - 3\n",
    "        \n",
    "        \n",
    "        X.fillna(self.medians, inplace=True)\n",
    "        \n",
    "        return X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "ff95fb13",
   "metadata": {},
   "outputs": [],
   "source": [
    "# построение новых признаков"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "11e1d608",
   "metadata": {},
   "outputs": [],
   "source": [
    "# заменяем A/B на 0/1\n",
    "binary_to_numbers = {'A': 0, 'B': 1}\n",
    "\n",
    "train_df['Ecology_2'] = train_df['Ecology_2'].replace(binary_to_numbers)\n",
    "train_df['Ecology_3'] = train_df['Ecology_3'].replace(binary_to_numbers)\n",
    "train_df['Shops_2'] = train_df['Shops_2'].replace(binary_to_numbers)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "8a00e237",
   "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>DistrictId</th>\n",
       "      <th>DistrictSize</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>27</td>\n",
       "      <td>851</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>652</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>23</td>\n",
       "      <td>565</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>6</td>\n",
       "      <td>511</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>9</td>\n",
       "      <td>294</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  DistrictId  DistrictSize\n",
       "0         27           851\n",
       "1          1           652\n",
       "2         23           565\n",
       "3          6           511\n",
       "4          9           294"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# добавляем размер района\n",
    "district_size = train_df['DistrictId'].value_counts().reset_index()\\\n",
    "                    .rename(columns={'index':'DistrictId', 'DistrictId':'DistrictSize'})\n",
    "\n",
    "district_size.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "a479ebce",
   "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>Id</th>\n",
       "      <th>DistrictId</th>\n",
       "      <th>Rooms</th>\n",
       "      <th>Square</th>\n",
       "      <th>LifeSquare</th>\n",
       "      <th>KitchenSquare</th>\n",
       "      <th>Floor</th>\n",
       "      <th>HouseFloor</th>\n",
       "      <th>HouseYear</th>\n",
       "      <th>Ecology_1</th>\n",
       "      <th>...</th>\n",
       "      <th>Social_2</th>\n",
       "      <th>Social_3</th>\n",
       "      <th>Helthcare_2</th>\n",
       "      <th>Shops_1</th>\n",
       "      <th>Shops_2</th>\n",
       "      <th>Price</th>\n",
       "      <th>HouseFloor_outlier</th>\n",
       "      <th>Meter_price</th>\n",
       "      <th>LifeSquare_nan</th>\n",
       "      <th>DistrictSize</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11809</td>\n",
       "      <td>27</td>\n",
       "      <td>3.0</td>\n",
       "      <td>115.027311</td>\n",
       "      <td>102.027311</td>\n",
       "      <td>10.0</td>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>2014</td>\n",
       "      <td>0.075424</td>\n",
       "      <td>...</td>\n",
       "      <td>3097</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>305018.871089</td>\n",
       "      <td>0</td>\n",
       "      <td>2651.708272</td>\n",
       "      <td>1</td>\n",
       "      <td>851</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3013</td>\n",
       "      <td>22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>39.832524</td>\n",
       "      <td>23.169223</td>\n",
       "      <td>8.0</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "      <td>1966</td>\n",
       "      <td>0.118537</td>\n",
       "      <td>...</td>\n",
       "      <td>6207</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>177734.553407</td>\n",
       "      <td>0</td>\n",
       "      <td>4462.045990</td>\n",
       "      <td>0</td>\n",
       "      <td>78</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8215</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>78.342215</td>\n",
       "      <td>47.671972</td>\n",
       "      <td>10.0</td>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>1988</td>\n",
       "      <td>0.025609</td>\n",
       "      <td>...</td>\n",
       "      <td>5261</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>282078.720850</td>\n",
       "      <td>0</td>\n",
       "      <td>3600.596698</td>\n",
       "      <td>0</td>\n",
       "      <td>652</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2352</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>40.409907</td>\n",
       "      <td>34.409907</td>\n",
       "      <td>3.0</td>\n",
       "      <td>10</td>\n",
       "      <td>22</td>\n",
       "      <td>1977</td>\n",
       "      <td>0.007122</td>\n",
       "      <td>...</td>\n",
       "      <td>264</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>168106.007630</td>\n",
       "      <td>0</td>\n",
       "      <td>4160.019670</td>\n",
       "      <td>1</td>\n",
       "      <td>652</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>13866</td>\n",
       "      <td>94</td>\n",
       "      <td>2.0</td>\n",
       "      <td>64.285067</td>\n",
       "      <td>38.562517</td>\n",
       "      <td>9.0</td>\n",
       "      <td>16</td>\n",
       "      <td>16</td>\n",
       "      <td>1972</td>\n",
       "      <td>0.282798</td>\n",
       "      <td>...</td>\n",
       "      <td>8667</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>343995.102962</td>\n",
       "      <td>0</td>\n",
       "      <td>5351.088794</td>\n",
       "      <td>0</td>\n",
       "      <td>89</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 23 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Id DistrictId  Rooms      Square  LifeSquare  KitchenSquare  Floor  \\\n",
       "0  11809         27    3.0  115.027311  102.027311           10.0      4   \n",
       "1   3013         22    1.0   39.832524   23.169223            8.0      7   \n",
       "2   8215          1    3.0   78.342215   47.671972           10.0      2   \n",
       "3   2352          1    1.0   40.409907   34.409907            3.0     10   \n",
       "4  13866         94    2.0   64.285067   38.562517            9.0     16   \n",
       "\n",
       "   HouseFloor  HouseYear  Ecology_1  ...  Social_2  Social_3  Helthcare_2  \\\n",
       "0          10       2014   0.075424  ...      3097         0            0   \n",
       "1           8       1966   0.118537  ...      6207         1            1   \n",
       "2          17       1988   0.025609  ...      5261         0            3   \n",
       "3          22       1977   0.007122  ...       264         0            0   \n",
       "4          16       1972   0.282798  ...      8667         2            0   \n",
       "\n",
       "   Shops_1  Shops_2          Price  HouseFloor_outlier  Meter_price  \\\n",
       "0        0        1  305018.871089                   0  2651.708272   \n",
       "1        0        1  177734.553407                   0  4462.045990   \n",
       "2        1        1  282078.720850                   0  3600.596698   \n",
       "3        1        1  168106.007630                   0  4160.019670   \n",
       "4        6        1  343995.102962                   0  5351.088794   \n",
       "\n",
       "   LifeSquare_nan  DistrictSize  \n",
       "0               1           851  \n",
       "1               0            78  \n",
       "2               0           652  \n",
       "3               1           652  \n",
       "4               0            89  \n",
       "\n",
       "[5 rows x 23 columns]"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# добавляем призак в датасет\n",
    "train_df = train_df.merge(district_size, on='DistrictId', how='left')\n",
    "train_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "aebec7b0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "True     5069\n",
       "False    4931\n",
       "Name: DistrictSize, dtype: int64"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# большие районы\n",
    "(train_df['DistrictSize'] > 100).value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "bdc9ecf1",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df['IsDistrictLarge'] = (train_df['DistrictSize'] > 100).astype(int)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "9bc15614",
   "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>DistrictId</th>\n",
       "      <th>Rooms</th>\n",
       "      <th>MedPriceByDistrict</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>142799.434052</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>176210.812334</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>300610.359425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>146735.671740</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>202706.090239</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  DistrictId  Rooms  MedPriceByDistrict\n",
       "0          0    1.0       142799.434052\n",
       "1          0    2.0       176210.812334\n",
       "2          0    3.0       300610.359425\n",
       "3          1    1.0       146735.671740\n",
       "4          1    2.0       202706.090239"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# медианная стоимость по району\n",
    "med_price_by_district = train_df.groupby(['DistrictId', 'Rooms'], as_index=False).agg({'Price':'median'})\\\n",
    "                            .rename(columns={'Price':'MedPriceByDistrict'})\n",
    "\n",
    "med_price_by_district.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "b6129f8e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(675, 3)"
      ]
     },
     "execution_count": 52,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "med_price_by_district.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "0d3d34ef",
   "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>Id</th>\n",
       "      <th>DistrictId</th>\n",
       "      <th>Rooms</th>\n",
       "      <th>Square</th>\n",
       "      <th>LifeSquare</th>\n",
       "      <th>KitchenSquare</th>\n",
       "      <th>Floor</th>\n",
       "      <th>HouseFloor</th>\n",
       "      <th>HouseYear</th>\n",
       "      <th>Ecology_1</th>\n",
       "      <th>...</th>\n",
       "      <th>Helthcare_2</th>\n",
       "      <th>Shops_1</th>\n",
       "      <th>Shops_2</th>\n",
       "      <th>Price</th>\n",
       "      <th>HouseFloor_outlier</th>\n",
       "      <th>Meter_price</th>\n",
       "      <th>LifeSquare_nan</th>\n",
       "      <th>DistrictSize</th>\n",
       "      <th>IsDistrictLarge</th>\n",
       "      <th>MedPriceByDistrict</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11809</td>\n",
       "      <td>27</td>\n",
       "      <td>3.0</td>\n",
       "      <td>115.027311</td>\n",
       "      <td>102.027311</td>\n",
       "      <td>10.0</td>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>2014</td>\n",
       "      <td>0.075424</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>305018.871089</td>\n",
       "      <td>0</td>\n",
       "      <td>2651.708272</td>\n",
       "      <td>1</td>\n",
       "      <td>851</td>\n",
       "      <td>1</td>\n",
       "      <td>213530.788443</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3013</td>\n",
       "      <td>22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>39.832524</td>\n",
       "      <td>23.169223</td>\n",
       "      <td>8.0</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "      <td>1966</td>\n",
       "      <td>0.118537</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>177734.553407</td>\n",
       "      <td>0</td>\n",
       "      <td>4462.045990</td>\n",
       "      <td>0</td>\n",
       "      <td>78</td>\n",
       "      <td>0</td>\n",
       "      <td>146032.829834</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8215</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>78.342215</td>\n",
       "      <td>47.671972</td>\n",
       "      <td>10.0</td>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>1988</td>\n",
       "      <td>0.025609</td>\n",
       "      <td>...</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>282078.720850</td>\n",
       "      <td>0</td>\n",
       "      <td>3600.596698</td>\n",
       "      <td>0</td>\n",
       "      <td>652</td>\n",
       "      <td>1</td>\n",
       "      <td>244005.272380</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2352</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>40.409907</td>\n",
       "      <td>34.409907</td>\n",
       "      <td>3.0</td>\n",
       "      <td>10</td>\n",
       "      <td>22</td>\n",
       "      <td>1977</td>\n",
       "      <td>0.007122</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>168106.007630</td>\n",
       "      <td>0</td>\n",
       "      <td>4160.019670</td>\n",
       "      <td>1</td>\n",
       "      <td>652</td>\n",
       "      <td>1</td>\n",
       "      <td>146735.671740</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>13866</td>\n",
       "      <td>94</td>\n",
       "      <td>2.0</td>\n",
       "      <td>64.285067</td>\n",
       "      <td>38.562517</td>\n",
       "      <td>9.0</td>\n",
       "      <td>16</td>\n",
       "      <td>16</td>\n",
       "      <td>1972</td>\n",
       "      <td>0.282798</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>343995.102962</td>\n",
       "      <td>0</td>\n",
       "      <td>5351.088794</td>\n",
       "      <td>0</td>\n",
       "      <td>89</td>\n",
       "      <td>0</td>\n",
       "      <td>201584.441255</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 25 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Id DistrictId  Rooms      Square  LifeSquare  KitchenSquare  Floor  \\\n",
       "0  11809         27    3.0  115.027311  102.027311           10.0      4   \n",
       "1   3013         22    1.0   39.832524   23.169223            8.0      7   \n",
       "2   8215          1    3.0   78.342215   47.671972           10.0      2   \n",
       "3   2352          1    1.0   40.409907   34.409907            3.0     10   \n",
       "4  13866         94    2.0   64.285067   38.562517            9.0     16   \n",
       "\n",
       "   HouseFloor  HouseYear  Ecology_1  ...  Helthcare_2  Shops_1  Shops_2  \\\n",
       "0          10       2014   0.075424  ...            0        0        1   \n",
       "1           8       1966   0.118537  ...            1        0        1   \n",
       "2          17       1988   0.025609  ...            3        1        1   \n",
       "3          22       1977   0.007122  ...            0        1        1   \n",
       "4          16       1972   0.282798  ...            0        6        1   \n",
       "\n",
       "           Price  HouseFloor_outlier  Meter_price  LifeSquare_nan  \\\n",
       "0  305018.871089                   0  2651.708272               1   \n",
       "1  177734.553407                   0  4462.045990               0   \n",
       "2  282078.720850                   0  3600.596698               0   \n",
       "3  168106.007630                   0  4160.019670               1   \n",
       "4  343995.102962                   0  5351.088794               0   \n",
       "\n",
       "   DistrictSize  IsDistrictLarge  MedPriceByDistrict  \n",
       "0           851                1       213530.788443  \n",
       "1            78                0       146032.829834  \n",
       "2           652                1       244005.272380  \n",
       "3           652                1       146735.671740  \n",
       "4            89                0       201584.441255  \n",
       "\n",
       "[5 rows x 25 columns]"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df = train_df.merge(med_price_by_district, on=['DistrictId', 'Rooms'], how='left')\n",
    "train_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "16892db7",
   "metadata": {},
   "outputs": [],
   "source": [
    "# категоризация\n",
    "def floor_to_cat(X):\n",
    "\n",
    "    X['floor_cat'] = 0\n",
    "\n",
    "    X.loc[X['Floor'] <= 3, 'floor_cat'] = 1  \n",
    "    X.loc[(X['Floor'] > 3) & (X['Floor'] <= 5), 'floor_cat'] = 2\n",
    "    X.loc[(X['Floor'] > 5) & (X['Floor'] <= 9), 'floor_cat'] = 3\n",
    "    X.loc[(X['Floor'] > 9) & (X['Floor'] <= 15), 'floor_cat'] = 4\n",
    "    X.loc[X['Floor'] > 15, 'floor_cat'] = 5\n",
    "\n",
    "    return X\n",
    "\n",
    "\n",
    "def floor_to_cat_pandas(X):\n",
    "    bins = [0, 3, 5, 9, 15, X['Floor'].max()]\n",
    "    X['floor_cat'] = pd.cut(X['Floor'], bins=bins, labels=False)\n",
    "    \n",
    "    X['floor_cat'].fillna(-1, inplace=True)\n",
    "    return X\n",
    "\n",
    "\n",
    "def year_to_cat(X):\n",
    "\n",
    "    X['year_cat'] = 0\n",
    "\n",
    "    X.loc[X['HouseYear'] <= 1941, 'year_cat'] = 1\n",
    "    X.loc[(X['HouseYear'] > 1941) & (X['HouseYear'] <= 1945), 'year_cat'] = 2\n",
    "    X.loc[(X['HouseYear'] > 1945) & (X['HouseYear'] <= 1980), 'year_cat'] = 3\n",
    "    X.loc[(X['HouseYear'] > 1980) & (X['HouseYear'] <= 2000), 'year_cat'] = 4\n",
    "    X.loc[(X['HouseYear'] > 2000) & (X['HouseYear'] <= 2010), 'year_cat'] = 5\n",
    "    X.loc[(X['HouseYear'] > 2010), 'year_cat'] = 6\n",
    "\n",
    "    return X\n",
    "\n",
    "\n",
    "def year_to_cat_pandas(X):\n",
    "    bins = [0, 1941, 1945, 1980, 2000, 2010, X['HouseYear'].max()]\n",
    "    X['year_cat'] = pd.cut(X['HouseYear'], bins=bins, labels=False)\n",
    "    \n",
    "    X['year_cat'].fillna(-1, inplace=True)\n",
    "    return X"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "95906706",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0       1\n",
       "1       2\n",
       "2       0\n",
       "3       3\n",
       "4       4\n",
       "       ..\n",
       "9995    0\n",
       "9996    3\n",
       "9997    0\n",
       "9998    0\n",
       "9999    2\n",
       "Name: Floor, Length: 10000, dtype: int64"
      ]
     },
     "execution_count": 55,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bins = [0, 3, 5, 9, 15, train_df['Floor'].max()]\n",
    "pd.cut(train_df['Floor'], bins=bins, labels=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "e47e6ba7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0         (3, 5]\n",
       "1         (5, 9]\n",
       "2         (0, 3]\n",
       "3        (9, 15]\n",
       "4       (15, 42]\n",
       "          ...   \n",
       "9995      (0, 3]\n",
       "9996     (9, 15]\n",
       "9997      (0, 3]\n",
       "9998      (0, 3]\n",
       "9999      (5, 9]\n",
       "Name: Floor, Length: 10000, dtype: category\n",
       "Categories (5, interval[int64, right]): [(0, 3] < (3, 5] < (5, 9] < (9, 15] < (15, 42]]"
      ]
     },
     "execution_count": 56,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "bins = [0, 3, 5, 9, 15, train_df['Floor'].max()]\n",
    "pd.cut(train_df['Floor'], bins=bins)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "20ca77d2",
   "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>Id</th>\n",
       "      <th>DistrictId</th>\n",
       "      <th>Rooms</th>\n",
       "      <th>Square</th>\n",
       "      <th>LifeSquare</th>\n",
       "      <th>KitchenSquare</th>\n",
       "      <th>Floor</th>\n",
       "      <th>HouseFloor</th>\n",
       "      <th>HouseYear</th>\n",
       "      <th>Ecology_1</th>\n",
       "      <th>...</th>\n",
       "      <th>Shops_2</th>\n",
       "      <th>Price</th>\n",
       "      <th>HouseFloor_outlier</th>\n",
       "      <th>Meter_price</th>\n",
       "      <th>LifeSquare_nan</th>\n",
       "      <th>DistrictSize</th>\n",
       "      <th>IsDistrictLarge</th>\n",
       "      <th>MedPriceByDistrict</th>\n",
       "      <th>year_cat</th>\n",
       "      <th>floor_cat</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11809</td>\n",
       "      <td>27</td>\n",
       "      <td>3.0</td>\n",
       "      <td>115.027311</td>\n",
       "      <td>102.027311</td>\n",
       "      <td>10.0</td>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>2014</td>\n",
       "      <td>0.075424</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>305018.871089</td>\n",
       "      <td>0</td>\n",
       "      <td>2651.708272</td>\n",
       "      <td>1</td>\n",
       "      <td>851</td>\n",
       "      <td>1</td>\n",
       "      <td>213530.788443</td>\n",
       "      <td>6</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3013</td>\n",
       "      <td>22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>39.832524</td>\n",
       "      <td>23.169223</td>\n",
       "      <td>8.0</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "      <td>1966</td>\n",
       "      <td>0.118537</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>177734.553407</td>\n",
       "      <td>0</td>\n",
       "      <td>4462.045990</td>\n",
       "      <td>0</td>\n",
       "      <td>78</td>\n",
       "      <td>0</td>\n",
       "      <td>146032.829834</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8215</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>78.342215</td>\n",
       "      <td>47.671972</td>\n",
       "      <td>10.0</td>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>1988</td>\n",
       "      <td>0.025609</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>282078.720850</td>\n",
       "      <td>0</td>\n",
       "      <td>3600.596698</td>\n",
       "      <td>0</td>\n",
       "      <td>652</td>\n",
       "      <td>1</td>\n",
       "      <td>244005.272380</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2352</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>40.409907</td>\n",
       "      <td>34.409907</td>\n",
       "      <td>3.0</td>\n",
       "      <td>10</td>\n",
       "      <td>22</td>\n",
       "      <td>1977</td>\n",
       "      <td>0.007122</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>168106.007630</td>\n",
       "      <td>0</td>\n",
       "      <td>4160.019670</td>\n",
       "      <td>1</td>\n",
       "      <td>652</td>\n",
       "      <td>1</td>\n",
       "      <td>146735.671740</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>13866</td>\n",
       "      <td>94</td>\n",
       "      <td>2.0</td>\n",
       "      <td>64.285067</td>\n",
       "      <td>38.562517</td>\n",
       "      <td>9.0</td>\n",
       "      <td>16</td>\n",
       "      <td>16</td>\n",
       "      <td>1972</td>\n",
       "      <td>0.282798</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>343995.102962</td>\n",
       "      <td>0</td>\n",
       "      <td>5351.088794</td>\n",
       "      <td>0</td>\n",
       "      <td>89</td>\n",
       "      <td>0</td>\n",
       "      <td>201584.441255</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 27 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Id DistrictId  Rooms      Square  LifeSquare  KitchenSquare  Floor  \\\n",
       "0  11809         27    3.0  115.027311  102.027311           10.0      4   \n",
       "1   3013         22    1.0   39.832524   23.169223            8.0      7   \n",
       "2   8215          1    3.0   78.342215   47.671972           10.0      2   \n",
       "3   2352          1    1.0   40.409907   34.409907            3.0     10   \n",
       "4  13866         94    2.0   64.285067   38.562517            9.0     16   \n",
       "\n",
       "   HouseFloor  HouseYear  Ecology_1  ...  Shops_2          Price  \\\n",
       "0          10       2014   0.075424  ...        1  305018.871089   \n",
       "1           8       1966   0.118537  ...        1  177734.553407   \n",
       "2          17       1988   0.025609  ...        1  282078.720850   \n",
       "3          22       1977   0.007122  ...        1  168106.007630   \n",
       "4          16       1972   0.282798  ...        1  343995.102962   \n",
       "\n",
       "   HouseFloor_outlier  Meter_price  LifeSquare_nan  DistrictSize  \\\n",
       "0                   0  2651.708272               1           851   \n",
       "1                   0  4462.045990               0            78   \n",
       "2                   0  3600.596698               0           652   \n",
       "3                   0  4160.019670               1           652   \n",
       "4                   0  5351.088794               0            89   \n",
       "\n",
       "   IsDistrictLarge  MedPriceByDistrict  year_cat  floor_cat  \n",
       "0                1       213530.788443         6          2  \n",
       "1                0       146032.829834         3          3  \n",
       "2                1       244005.272380         4          1  \n",
       "3                1       146735.671740         3          4  \n",
       "4                0       201584.441255         3          5  \n",
       "\n",
       "[5 rows x 27 columns]"
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df = year_to_cat(train_df)\n",
    "train_df = floor_to_cat(train_df)\n",
    "train_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "a3fb003c",
   "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>year_cat</th>\n",
       "      <th>floor_cat</th>\n",
       "      <th>MedPriceByFloorYear</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>374737.658654</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>322905.044793</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>429825.896082</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>274992.472366</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>467230.539057</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   year_cat  floor_cat  MedPriceByFloorYear\n",
       "0         1          1        374737.658654\n",
       "1         1          2        322905.044793\n",
       "2         1          3        429825.896082\n",
       "3         1          4        274992.472366\n",
       "4         2          1        467230.539057"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#!!! медианная цена по году и этажу\n",
    "med_price_by_floor_year = train_df.groupby(['year_cat', 'floor_cat'], as_index=False).agg({'Price':'median'}).\\\n",
    "                                            rename(columns={'Price':'MedPriceByFloorYear'})\n",
    "med_price_by_floor_year.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "16adca39",
   "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>Id</th>\n",
       "      <th>DistrictId</th>\n",
       "      <th>Rooms</th>\n",
       "      <th>Square</th>\n",
       "      <th>LifeSquare</th>\n",
       "      <th>KitchenSquare</th>\n",
       "      <th>Floor</th>\n",
       "      <th>HouseFloor</th>\n",
       "      <th>HouseYear</th>\n",
       "      <th>Ecology_1</th>\n",
       "      <th>...</th>\n",
       "      <th>Price</th>\n",
       "      <th>HouseFloor_outlier</th>\n",
       "      <th>Meter_price</th>\n",
       "      <th>LifeSquare_nan</th>\n",
       "      <th>DistrictSize</th>\n",
       "      <th>IsDistrictLarge</th>\n",
       "      <th>MedPriceByDistrict</th>\n",
       "      <th>year_cat</th>\n",
       "      <th>floor_cat</th>\n",
       "      <th>MedPriceByFloorYear</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11809</td>\n",
       "      <td>27</td>\n",
       "      <td>3.0</td>\n",
       "      <td>115.027311</td>\n",
       "      <td>102.027311</td>\n",
       "      <td>10.0</td>\n",
       "      <td>4</td>\n",
       "      <td>10</td>\n",
       "      <td>2014</td>\n",
       "      <td>0.075424</td>\n",
       "      <td>...</td>\n",
       "      <td>305018.871089</td>\n",
       "      <td>0</td>\n",
       "      <td>2651.708272</td>\n",
       "      <td>1</td>\n",
       "      <td>851</td>\n",
       "      <td>1</td>\n",
       "      <td>213530.788443</td>\n",
       "      <td>6</td>\n",
       "      <td>2</td>\n",
       "      <td>161974.797753</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3013</td>\n",
       "      <td>22</td>\n",
       "      <td>1.0</td>\n",
       "      <td>39.832524</td>\n",
       "      <td>23.169223</td>\n",
       "      <td>8.0</td>\n",
       "      <td>7</td>\n",
       "      <td>8</td>\n",
       "      <td>1966</td>\n",
       "      <td>0.118537</td>\n",
       "      <td>...</td>\n",
       "      <td>177734.553407</td>\n",
       "      <td>0</td>\n",
       "      <td>4462.045990</td>\n",
       "      <td>0</td>\n",
       "      <td>78</td>\n",
       "      <td>0</td>\n",
       "      <td>146032.829834</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>192538.872015</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>8215</td>\n",
       "      <td>1</td>\n",
       "      <td>3.0</td>\n",
       "      <td>78.342215</td>\n",
       "      <td>47.671972</td>\n",
       "      <td>10.0</td>\n",
       "      <td>2</td>\n",
       "      <td>17</td>\n",
       "      <td>1988</td>\n",
       "      <td>0.025609</td>\n",
       "      <td>...</td>\n",
       "      <td>282078.720850</td>\n",
       "      <td>0</td>\n",
       "      <td>3600.596698</td>\n",
       "      <td>0</td>\n",
       "      <td>652</td>\n",
       "      <td>1</td>\n",
       "      <td>244005.272380</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>204670.096676</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2352</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>40.409907</td>\n",
       "      <td>34.409907</td>\n",
       "      <td>3.0</td>\n",
       "      <td>10</td>\n",
       "      <td>22</td>\n",
       "      <td>1977</td>\n",
       "      <td>0.007122</td>\n",
       "      <td>...</td>\n",
       "      <td>168106.007630</td>\n",
       "      <td>0</td>\n",
       "      <td>4160.019670</td>\n",
       "      <td>1</td>\n",
       "      <td>652</td>\n",
       "      <td>1</td>\n",
       "      <td>146735.671740</td>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "      <td>183505.979560</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>13866</td>\n",
       "      <td>94</td>\n",
       "      <td>2.0</td>\n",
       "      <td>64.285067</td>\n",
       "      <td>38.562517</td>\n",
       "      <td>9.0</td>\n",
       "      <td>16</td>\n",
       "      <td>16</td>\n",
       "      <td>1972</td>\n",
       "      <td>0.282798</td>\n",
       "      <td>...</td>\n",
       "      <td>343995.102962</td>\n",
       "      <td>0</td>\n",
       "      <td>5351.088794</td>\n",
       "      <td>0</td>\n",
       "      <td>89</td>\n",
       "      <td>0</td>\n",
       "      <td>201584.441255</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>173955.556579</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 28 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      Id DistrictId  Rooms      Square  LifeSquare  KitchenSquare  Floor  \\\n",
       "0  11809         27    3.0  115.027311  102.027311           10.0      4   \n",
       "1   3013         22    1.0   39.832524   23.169223            8.0      7   \n",
       "2   8215          1    3.0   78.342215   47.671972           10.0      2   \n",
       "3   2352          1    1.0   40.409907   34.409907            3.0     10   \n",
       "4  13866         94    2.0   64.285067   38.562517            9.0     16   \n",
       "\n",
       "   HouseFloor  HouseYear  Ecology_1  ...          Price  HouseFloor_outlier  \\\n",
       "0          10       2014   0.075424  ...  305018.871089                   0   \n",
       "1           8       1966   0.118537  ...  177734.553407                   0   \n",
       "2          17       1988   0.025609  ...  282078.720850                   0   \n",
       "3          22       1977   0.007122  ...  168106.007630                   0   \n",
       "4          16       1972   0.282798  ...  343995.102962                   0   \n",
       "\n",
       "   Meter_price  LifeSquare_nan  DistrictSize  IsDistrictLarge  \\\n",
       "0  2651.708272               1           851                1   \n",
       "1  4462.045990               0            78                0   \n",
       "2  3600.596698               0           652                1   \n",
       "3  4160.019670               1           652                1   \n",
       "4  5351.088794               0            89                0   \n",
       "\n",
       "   MedPriceByDistrict  year_cat  floor_cat  MedPriceByFloorYear  \n",
       "0       213530.788443         6          2        161974.797753  \n",
       "1       146032.829834         3          3        192538.872015  \n",
       "2       244005.272380         4          1        204670.096676  \n",
       "3       146735.671740         3          4        183505.979560  \n",
       "4       201584.441255         3          5        173955.556579  \n",
       "\n",
       "[5 rows x 28 columns]"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_df = train_df.merge(med_price_by_floor_year, on=['year_cat', 'floor_cat'], how='left')\n",
    "train_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "1fa55c19",
   "metadata": {},
   "outputs": [],
   "source": [
    "class FeatureGenetator():\n",
    "    \"\"\"Генерация новых фич\"\"\"\n",
    "    \n",
    "    def __init__(self):\n",
    "        self.DistrictId_counts = None\n",
    "        self.binary_to_numbers = None\n",
    "        self.med_price_by_district = None\n",
    "        self.med_price_by_floor_year = None\n",
    "        self.house_year_max = None\n",
    "        self.floor_max = None\n",
    "        self.district_size = None\n",
    "        self.meter_price = None      #AV\n",
    "        self.meter_price_by_district_rooms = None   #AV\n",
    "        \n",
    "    def fit(self, X, y=None):\n",
    "        \n",
    "        X = X.copy()\n",
    "        \n",
    "        # Binary features\n",
    "        self.binary_to_numbers = {'A': 0, 'B': 1}\n",
    "        \n",
    "        # DistrictID\n",
    "        self.district_size = X['DistrictId'].value_counts().reset_index() \\\n",
    "                               .rename(columns={'index':'DistrictId', 'DistrictId':'DistrictSize'})\n",
    "                \n",
    "        # Target encoding\n",
    "        ## District, Rooms\n",
    "        df = X.copy()\n",
    "        \n",
    "        if y is not None:\n",
    "            df['Price'] = y.values\n",
    "            \n",
    "            self.med_price_by_district = df.groupby(['DistrictId', 'Rooms'], as_index=False).agg({'Price':'median'})\\\n",
    "                                            .rename(columns={'Price':'MedPriceByDistrict'})\n",
    "            \n",
    "            self.med_price_by_district_median = self.med_price_by_district['MedPriceByDistrict'].median()\n",
    "            \n",
    "        ## floor, year\n",
    "        if y is not None:\n",
    "            self.floor_max = df['Floor'].max()\n",
    "            self.house_year_max = df['HouseYear'].max()\n",
    "            df['Price'] = y.values\n",
    "            df = self.floor_to_cat(df)\n",
    "            df = self.year_to_cat(df)\n",
    "            self.med_price_by_floor_year = df.groupby(['year_cat', 'floor_cat'], as_index=False).agg({'Price':'median'}).\\\n",
    "                                            rename(columns={'Price':'MedPriceByFloorYear'})\n",
    "            self.med_price_by_floor_year_median = self.med_price_by_floor_year['MedPriceByFloorYear'].median()\n",
    "            \n",
    "        if y is not None:\n",
    "            df['Price'] = y.values\n",
    "            self.meter_price = df['Price']/df['Square']      #AV    \n",
    "        \n",
    "        \n",
    "        if y is not None:\n",
    "            df['Price'] = y.values            \n",
    "            df = self.fill_meter_price(df)\n",
    "            self.meter_price_by_district_rooms = df.groupby(['DistrictId', 'Rooms'], as_index=False).\\\n",
    "                                            agg({'MeterPrice':'median'}).\\\n",
    "                                            rename(columns={'MeterPrice':'MeterPriceByDistrictRooms'})\n",
    "            \n",
    "          \n",
    "        \n",
    "    def transform(self, X):\n",
    "        \n",
    "        # Binary features\n",
    "        X['Ecology_2'] = X['Ecology_2'].map(self.binary_to_numbers)  # self.binary_to_numbers = {'A': 0, 'B': 1}\n",
    "        X['Ecology_3'] = X['Ecology_3'].map(self.binary_to_numbers)\n",
    "        X['Shops_2'] = X['Shops_2'].map(self.binary_to_numbers)\n",
    "        \n",
    "        # DistrictId, IsDistrictLarge\n",
    "        X = X.merge(self.district_size, on='DistrictId', how='left')\n",
    "        \n",
    "        X['new_district'] = 0\n",
    "        X.loc[X['DistrictSize'].isna(), 'new_district'] = 1\n",
    "        \n",
    "        X['DistrictSize'].fillna(5, inplace=True)\n",
    "        \n",
    "        X['IsDistrictLarge'] = (X['DistrictSize'] > 100).astype(int)\n",
    "        \n",
    "        # More categorical features\n",
    "        X = self.floor_to_cat(X)  # + столбец floor_cat\n",
    "        X = self.year_to_cat(X)   # + столбец year_cat\n",
    "        \n",
    "        # Target encoding\n",
    "        if self.med_price_by_district is not None:\n",
    "            X = X.merge(self.med_price_by_district, on=['DistrictId', 'Rooms'], how='left')\n",
    "            X['MedPriceByDistrict'].fillna(self.med_price_by_district_median, inplace=True)\n",
    "            \n",
    "        if self.med_price_by_floor_year is not None:\n",
    "            X = X.merge(self.med_price_by_floor_year, on=['year_cat', 'floor_cat'], how='left')\n",
    "            X['MedPriceByFloorYear'].fillna(self.med_price_by_floor_year_median, inplace=True)\n",
    "          \n",
    "        # AV\n",
    "        X['MeterPrice'] = self.meter_price\n",
    "        X['MeterPrice'].fillna(3910,inplace = True)\n",
    "        #X.loc[(X['Square'] > 0), 'MeterPrice'] = X['Price'] / X['Square'] \n",
    "        \n",
    "        if self.meter_price_by_district_rooms is not None:\n",
    "            X = X.merge(self.meter_price_by_district_rooms, on=['DistrictId', 'Rooms'], how='left')\n",
    "            X['MeterPriceByDistrictRooms'].fillna(self.med_price_by_floor_year_median, inplace=True)        \n",
    "        \n",
    "                \n",
    "        return X    \n",
    "     \n",
    "    def floor_to_cat(self, X):\n",
    "        bins = [0, 3, 5, 9, 15, self.floor_max]\n",
    "        X['floor_cat'] = pd.cut(X['Floor'], bins=bins, labels=False)\n",
    "\n",
    "        X['floor_cat'].fillna(-1, inplace=True)\n",
    "        return X\n",
    "     \n",
    "    def year_to_cat(self, X):\n",
    "        bins = [0, 1941, 1945, 1980, 2000, 2010, self.house_year_max]\n",
    "        X['year_cat'] = pd.cut(X['HouseYear'], bins=bins, labels=False)\n",
    "        X['year_cat'].fillna(-1, inplace=True)\n",
    "        return X\n",
    "    \n",
    "    def fill_meter_price(self, X):    #AV\n",
    "        X['MeterPrice'] = self.meter_price\n",
    "        X['MeterPrice'].fillna(3910,inplace = True)\n",
    "        X.loc[(X['Square'] > 0), 'MeterPrice'] = X['Price'] / X['Square']\n",
    "        return X\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "6e21b269",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['Id',\n",
       " 'DistrictId',\n",
       " 'Rooms',\n",
       " 'Square',\n",
       " 'LifeSquare',\n",
       " 'KitchenSquare',\n",
       " 'Floor',\n",
       " 'HouseFloor',\n",
       " 'HouseYear',\n",
       " 'Ecology_1',\n",
       " 'Ecology_2',\n",
       " 'Ecology_3',\n",
       " 'Social_1',\n",
       " 'Social_2',\n",
       " 'Social_3',\n",
       " 'Helthcare_2',\n",
       " 'Shops_1',\n",
       " 'Shops_2',\n",
       " 'Price',\n",
       " 'HouseFloor_outlier',\n",
       " 'Meter_price',\n",
       " 'LifeSquare_nan',\n",
       " 'DistrictSize',\n",
       " 'IsDistrictLarge',\n",
       " 'MedPriceByDistrict',\n",
       " 'year_cat',\n",
       " 'floor_cat',\n",
       " 'MedPriceByFloorYear']"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# отбор призаков\n",
    "train_df.columns.tolist()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "b70cf62f",
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_names = ['Rooms', 'Square', 'LifeSquare', 'KitchenSquare', 'Floor', 'HouseFloor', 'HouseYear',\n",
    "                 'Ecology_1', 'Ecology_2', 'Ecology_3', 'Social_1', 'Social_2', 'Social_3',\n",
    "                 'Helthcare_2', 'Shops_1', 'Shops_2']\n",
    "\n",
    "new_feature_names = ['Rooms_outlier', 'HouseFloor_outlier', 'HouseYear_outlier', 'LifeSquare_nan', 'DistrictSize',\n",
    "                     'new_district', 'IsDistrictLarge',  'MedPriceByDistrict', 'MedPriceByFloorYear', 'MeterPrice',\n",
    "                     'MeterPriceByDistrictRooms']\n",
    "\n",
    "target_name = 'Price'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "6cdc1577",
   "metadata": {},
   "outputs": [],
   "source": [
    "# разбиение на train и test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "31fa43e6",
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df = pd.read_csv(TRAIN_DATASET_PATH)\n",
    "test_df = pd.read_csv(TEST_DATASET_PATH)\n",
    "\n",
    "X = train_df.drop(columns=target_name)\n",
    "y = train_df[target_name]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "0fac92ea",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_valid, y_train, y_valid = train_test_split(X, y, test_size=0.33, shuffle=True, random_state=21)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "3209cde6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((6700, 22), (3300, 22), (5000, 22))"
      ]
     },
     "execution_count": 66,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "preprocessor = DataPreprocessing()\n",
    "preprocessor.fit(X_train)\n",
    "\n",
    "X_train = preprocessor.transform(X_train)\n",
    "X_valid = preprocessor.transform(X_valid)\n",
    "test_df = preprocessor.transform(test_df)\n",
    "\n",
    "X_train.shape, X_valid.shape, test_df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "f9474705",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((6700, 31), (3300, 31), (5000, 31))"
      ]
     },
     "execution_count": 67,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "features_gen = FeatureGenetator()\n",
    "features_gen.fit(X_train, y_train)\n",
    "\n",
    "X_train = features_gen.transform(X_train)\n",
    "X_valid = features_gen.transform(X_valid)\n",
    "test_df = features_gen.transform(test_df)\n",
    "\n",
    "X_train.shape, X_valid.shape, test_df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "6fba65c6",
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train = X_train[feature_names + new_feature_names]\n",
    "X_valid = X_valid[feature_names + new_feature_names]\n",
    "test_df = test_df[feature_names + new_feature_names]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "a67cd879",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0, 0, 0)"
      ]
     },
     "execution_count": 69,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "X_train.isna().sum().sum(), X_valid.isna().sum().sum(), test_df.isna().sum().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "4c9c6e00",
   "metadata": {},
   "outputs": [],
   "source": [
    "# построение модели"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "48adceb1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(random_state=21)"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "rf_model = RandomForestRegressor(random_state=21, criterion='mse')\n",
    "rf_model.fit(X_train, y_train)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "984202de",
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Train R2:\t0.967\n",
      "Test R2:\t0.722\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x720 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    " # оценка модели\n",
    "y_train_preds = rf_model.predict(X_train)\n",
    "y_test_preds = rf_model.predict(X_valid)\n",
    "\n",
    "evaluate_preds(y_train, y_train_preds, y_valid, y_test_preds)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "f880df39",
   "metadata": {},
   "outputs": [],
   "source": [
    "# кросс валидация"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "0290cf24",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([0.78324412, 0.73308622, 0.76224678])"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cv_score = cross_val_score(rf_model, X_train, y_train, scoring='r2', cv=KFold(n_splits=3, shuffle=True, random_state=21))\n",
    "cv_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "5d373cf1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.7595257056684376"
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cv_score.mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "b5be4638",
   "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>feature_name</th>\n",
       "      <th>importance</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>MedPriceByDistrict</td>\n",
       "      <td>6.377276e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Square</td>\n",
       "      <td>1.169206e-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>MeterPriceByDistrictRooms</td>\n",
       "      <td>5.302766e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>LifeSquare</td>\n",
       "      <td>2.742734e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>HouseYear</td>\n",
       "      <td>1.880421e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>MedPriceByFloorYear</td>\n",
       "      <td>1.836782e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>HouseFloor</td>\n",
       "      <td>1.637686e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>MeterPrice</td>\n",
       "      <td>1.623973e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Floor</td>\n",
       "      <td>1.337890e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>KitchenSquare</td>\n",
       "      <td>1.172162e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Social_3</td>\n",
       "      <td>1.140762e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Ecology_1</td>\n",
       "      <td>1.114983e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Social_2</td>\n",
       "      <td>1.038791e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>DistrictSize</td>\n",
       "      <td>1.014266e-02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Social_1</td>\n",
       "      <td>9.488800e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Shops_1</td>\n",
       "      <td>7.056769e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Helthcare_2</td>\n",
       "      <td>4.669791e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>HouseFloor_outlier</td>\n",
       "      <td>1.497970e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Rooms</td>\n",
       "      <td>1.197059e-03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>LifeSquare_nan</td>\n",
       "      <td>8.821891e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Shops_2</td>\n",
       "      <td>8.546742e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Ecology_3</td>\n",
       "      <td>6.578909e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>IsDistrictLarge</td>\n",
       "      <td>4.151446e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Ecology_2</td>\n",
       "      <td>1.322893e-04</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Rooms_outlier</td>\n",
       "      <td>6.708606e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>HouseYear_outlier</td>\n",
       "      <td>1.218268e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>new_district</td>\n",
       "      <td>0.000000e+00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 feature_name    importance\n",
       "23         MedPriceByDistrict  6.377276e-01\n",
       "1                      Square  1.169206e-01\n",
       "26  MeterPriceByDistrictRooms  5.302766e-02\n",
       "2                  LifeSquare  2.742734e-02\n",
       "6                   HouseYear  1.880421e-02\n",
       "24        MedPriceByFloorYear  1.836782e-02\n",
       "5                  HouseFloor  1.637686e-02\n",
       "25                 MeterPrice  1.623973e-02\n",
       "4                       Floor  1.337890e-02\n",
       "3               KitchenSquare  1.172162e-02\n",
       "12                   Social_3  1.140762e-02\n",
       "7                   Ecology_1  1.114983e-02\n",
       "11                   Social_2  1.038791e-02\n",
       "20               DistrictSize  1.014266e-02\n",
       "10                   Social_1  9.488800e-03\n",
       "14                    Shops_1  7.056769e-03\n",
       "13                Helthcare_2  4.669791e-03\n",
       "17         HouseFloor_outlier  1.497970e-03\n",
       "0                       Rooms  1.197059e-03\n",
       "19             LifeSquare_nan  8.821891e-04\n",
       "15                    Shops_2  8.546742e-04\n",
       "9                   Ecology_3  6.578909e-04\n",
       "22            IsDistrictLarge  4.151446e-04\n",
       "8                   Ecology_2  1.322893e-04\n",
       "16              Rooms_outlier  6.708606e-05\n",
       "18          HouseYear_outlier  1.218268e-08\n",
       "21               new_district  0.000000e+00"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# веса признаков\n",
    "feature_importances = pd.DataFrame(zip(X_train.columns, rf_model.feature_importances_), \n",
    "                                   columns=['feature_name', 'importance'])\n",
    "\n",
    "feature_importances.sort_values(by='importance', ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "7d95cb34",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(5000, 27)"
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# выгрузка результатов\n",
    "test_df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "id": "90e14b5d",
   "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>Id</th>\n",
       "      <th>Price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4567</td>\n",
       "      <td>200000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5925</td>\n",
       "      <td>200000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>960</td>\n",
       "      <td>200000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3848</td>\n",
       "      <td>200000.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>746</td>\n",
       "      <td>200000.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Id     Price\n",
       "0  4567  200000.0\n",
       "1  5925  200000.0\n",
       "2   960  200000.0\n",
       "3  3848  200000.0\n",
       "4   746  200000.0"
      ]
     },
     "execution_count": 78,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "submit = pd.read_csv('C:/ARTEM/GeekBrains/Python4DS/RealEstatePricePredictionMoscow/sample_submission.csv')\n",
    "submit.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "id": "2c01f0a4",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([164129.41193348, 117169.5995882 , 139193.78182605, ...,\n",
       "       148103.70058817, 189761.30723796, 229074.87215388])"
      ]
     },
     "execution_count": 79,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "predictions = rf_model.predict(test_df)\n",
    "predictions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "id": "6c9f1523",
   "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>Id</th>\n",
       "      <th>Price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>4567</td>\n",
       "      <td>164129.411933</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>5925</td>\n",
       "      <td>117169.599588</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>960</td>\n",
       "      <td>139193.781826</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3848</td>\n",
       "      <td>169578.024348</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>746</td>\n",
       "      <td>171922.946686</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     Id          Price\n",
       "0  4567  164129.411933\n",
       "1  5925  117169.599588\n",
       "2   960  139193.781826\n",
       "3  3848  169578.024348\n",
       "4   746  171922.946686"
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "submit['Price'] = predictions\n",
    "submit.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "id": "e6052d38",
   "metadata": {},
   "outputs": [],
   "source": [
    "submit.to_csv('C:/ARTEM/GeekBrains/Python4DS/RealEstatePricePredictionMoscow/rf_submit.csv', index=False)"
   ]
  }
 ],
 "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": "3.9.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}