{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install ЧегоНеХватает"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.7/site-packages/lightgbm/__init__.py:46: UserWarning: Starting from version 2.2.1, the library file in distribution wheels for macOS is built by the Apple Clang (Xcode_8.3.1) compiler.\n",
      "This means that in case of installing LightGBM from PyPI via the ``pip install lightgbm`` command, you don't need to install the gcc compiler anymore.\n",
      "Instead of that, you need to install the OpenMP library, which is required for running LightGBM on the system with the Apple Clang compiler.\n",
      "You can install the OpenMP library by the following command: ``brew install libomp``.\n",
      "  \"You can install the OpenMP library by the following command: ``brew install libomp``.\", UserWarning)\n"
     ]
    }
   ],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import lightgbm # сожрет все сырым и построит регрессионную модель, которая покажет важные фичи\n",
    "                # чтобы дальше делать лабу только на них\n",
    "                # если не можешь установить, то важные фичи в ячейке [8]\n",
    "from sklearn.model_selection import train_test_split, GridSearchCV\n",
    "from sklearn.neighbors import KNeighborsRegressor"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "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>bulk_id</th>\n",
       "      <th>spalen</th>\n",
       "      <th>date1</th>\n",
       "      <th>start_square</th>\n",
       "      <th>value</th>\n",
       "      <th>price</th>\n",
       "      <th>mean_sq</th>\n",
       "      <th>mean_fl</th>\n",
       "      <th>plan_s</th>\n",
       "      <th>plan_m</th>\n",
       "      <th>...</th>\n",
       "      <th>До промки(км)</th>\n",
       "      <th>До парка(км)</th>\n",
       "      <th>До парка пешком(км)</th>\n",
       "      <th>Станций метро от кольца</th>\n",
       "      <th>Площадь двора</th>\n",
       "      <th>Курс</th>\n",
       "      <th>Cтавка по ипотеке</th>\n",
       "      <th>Вклады до 1 года</th>\n",
       "      <th>Вклады от 1 года до 3 лет</th>\n",
       "      <th>Вклады свыше 3 лет</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>FF3814A9-9F7D-E711-8530-00505688958B</td>\n",
       "      <td>1</td>\n",
       "      <td>2017-10-01</td>\n",
       "      <td>13109.9</td>\n",
       "      <td>1124.5</td>\n",
       "      <td>135633.913043</td>\n",
       "      <td>38.00</td>\n",
       "      <td>8</td>\n",
       "      <td>147</td>\n",
       "      <td>198</td>\n",
       "      <td>...</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.93</td>\n",
       "      <td>8</td>\n",
       "      <td>7640</td>\n",
       "      <td>57.698052</td>\n",
       "      <td>10.04</td>\n",
       "      <td>5.74</td>\n",
       "      <td>5.94</td>\n",
       "      <td>6.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>FF3814A9-9F7D-E711-8530-00505688958B</td>\n",
       "      <td>2</td>\n",
       "      <td>2017-10-01</td>\n",
       "      <td>19898.4</td>\n",
       "      <td>1412.3</td>\n",
       "      <td>128492.419825</td>\n",
       "      <td>58.01</td>\n",
       "      <td>9</td>\n",
       "      <td>139</td>\n",
       "      <td>120</td>\n",
       "      <td>...</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.93</td>\n",
       "      <td>8</td>\n",
       "      <td>7640</td>\n",
       "      <td>57.698052</td>\n",
       "      <td>10.04</td>\n",
       "      <td>5.74</td>\n",
       "      <td>5.94</td>\n",
       "      <td>6.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>FF3814A9-9F7D-E711-8530-00505688958B</td>\n",
       "      <td>0</td>\n",
       "      <td>2017-10-01</td>\n",
       "      <td>2100.0</td>\n",
       "      <td>345.0</td>\n",
       "      <td>158237.634409</td>\n",
       "      <td>22.58</td>\n",
       "      <td>10</td>\n",
       "      <td>78</td>\n",
       "      <td>15</td>\n",
       "      <td>...</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.93</td>\n",
       "      <td>8</td>\n",
       "      <td>7640</td>\n",
       "      <td>57.698052</td>\n",
       "      <td>10.04</td>\n",
       "      <td>5.74</td>\n",
       "      <td>5.94</td>\n",
       "      <td>6.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>FF3814A9-9F7D-E711-8530-00505688958B</td>\n",
       "      <td>3</td>\n",
       "      <td>2017-10-01</td>\n",
       "      <td>9614.9</td>\n",
       "      <td>769.4</td>\n",
       "      <td>123250.000000</td>\n",
       "      <td>78.81</td>\n",
       "      <td>8</td>\n",
       "      <td>50</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.93</td>\n",
       "      <td>8</td>\n",
       "      <td>7640</td>\n",
       "      <td>57.698052</td>\n",
       "      <td>10.04</td>\n",
       "      <td>5.74</td>\n",
       "      <td>5.94</td>\n",
       "      <td>6.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>FF3814A9-9F7D-E711-8530-00505688958B</td>\n",
       "      <td>1</td>\n",
       "      <td>2017-11-01</td>\n",
       "      <td>11947.2</td>\n",
       "      <td>526.7</td>\n",
       "      <td>139420.634921</td>\n",
       "      <td>37.93</td>\n",
       "      <td>9</td>\n",
       "      <td>140</td>\n",
       "      <td>175</td>\n",
       "      <td>...</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.93</td>\n",
       "      <td>8</td>\n",
       "      <td>7640</td>\n",
       "      <td>58.926552</td>\n",
       "      <td>9.87</td>\n",
       "      <td>5.17</td>\n",
       "      <td>7.07</td>\n",
       "      <td>6.20</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 55 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 bulk_id  spalen       date1  start_square  \\\n",
       "id                                                                           \n",
       "0   FF3814A9-9F7D-E711-8530-00505688958B       1  2017-10-01       13109.9   \n",
       "1   FF3814A9-9F7D-E711-8530-00505688958B       2  2017-10-01       19898.4   \n",
       "2   FF3814A9-9F7D-E711-8530-00505688958B       0  2017-10-01        2100.0   \n",
       "3   FF3814A9-9F7D-E711-8530-00505688958B       3  2017-10-01        9614.9   \n",
       "4   FF3814A9-9F7D-E711-8530-00505688958B       1  2017-11-01       11947.2   \n",
       "\n",
       "     value          price  mean_sq  mean_fl  plan_s  plan_m  \\\n",
       "id                                                            \n",
       "0   1124.5  135633.913043    38.00        8     147     198   \n",
       "1   1412.3  128492.419825    58.01        9     139     120   \n",
       "2    345.0  158237.634409    22.58       10      78      15   \n",
       "3    769.4  123250.000000    78.81        8      50       0   \n",
       "4    526.7  139420.634921    37.93        9     140     175   \n",
       "\n",
       "           ...          До промки(км)  До парка(км)  До парка пешком(км)  \\\n",
       "id         ...                                                             \n",
       "0          ...                   0.28          0.58                 0.93   \n",
       "1          ...                   0.28          0.58                 0.93   \n",
       "2          ...                   0.28          0.58                 0.93   \n",
       "3          ...                   0.28          0.58                 0.93   \n",
       "4          ...                   0.28          0.58                 0.93   \n",
       "\n",
       "    Станций метро от кольца  Площадь двора       Курс Cтавка по ипотеке  \\\n",
       "id                                                                        \n",
       "0                         8           7640  57.698052             10.04   \n",
       "1                         8           7640  57.698052             10.04   \n",
       "2                         8           7640  57.698052             10.04   \n",
       "3                         8           7640  57.698052             10.04   \n",
       "4                         8           7640  58.926552              9.87   \n",
       "\n",
       "    Вклады до 1 года Вклады от 1 года до 3 лет  Вклады свыше 3 лет  \n",
       "id                                                                  \n",
       "0               5.74                      5.94                6.03  \n",
       "1               5.74                      5.94                6.03  \n",
       "2               5.74                      5.94                6.03  \n",
       "3               5.74                      5.94                6.03  \n",
       "4               5.17                      7.07                6.20  \n",
       "\n",
       "[5 rows x 55 columns]"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = pd.read_csv('PIK_Property.csv', encoding='cp1251', index_col=0)\n",
    "df.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "целевая переменная value"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "тут нет пропущенных значений, сделаем сами:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>bulk_id</th>\n",
       "      <th>spalen</th>\n",
       "      <th>date1</th>\n",
       "      <th>start_square</th>\n",
       "      <th>value</th>\n",
       "      <th>price</th>\n",
       "      <th>mean_sq</th>\n",
       "      <th>mean_fl</th>\n",
       "      <th>plan_s</th>\n",
       "      <th>plan_m</th>\n",
       "      <th>...</th>\n",
       "      <th>До промки(км)</th>\n",
       "      <th>До парка(км)</th>\n",
       "      <th>До парка пешком(км)</th>\n",
       "      <th>Станций метро от кольца</th>\n",
       "      <th>Площадь двора</th>\n",
       "      <th>Курс</th>\n",
       "      <th>Cтавка по ипотеке</th>\n",
       "      <th>Вклады до 1 года</th>\n",
       "      <th>Вклады от 1 года до 3 лет</th>\n",
       "      <th>Вклады свыше 3 лет</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>2017-10-01</td>\n",
       "      <td>13109.9</td>\n",
       "      <td>1124.5</td>\n",
       "      <td>135633.913043</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8</td>\n",
       "      <td>147</td>\n",
       "      <td>198</td>\n",
       "      <td>...</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.93</td>\n",
       "      <td>8</td>\n",
       "      <td>7640</td>\n",
       "      <td>57.698052</td>\n",
       "      <td>10.04</td>\n",
       "      <td>5.74</td>\n",
       "      <td>5.94</td>\n",
       "      <td>6.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>FF3814A9-9F7D-E711-8530-00505688958B</td>\n",
       "      <td>2</td>\n",
       "      <td>2017-10-01</td>\n",
       "      <td>19898.4</td>\n",
       "      <td>1412.3</td>\n",
       "      <td>128492.419825</td>\n",
       "      <td>58.01</td>\n",
       "      <td>9</td>\n",
       "      <td>139</td>\n",
       "      <td>120</td>\n",
       "      <td>...</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.93</td>\n",
       "      <td>8</td>\n",
       "      <td>7640</td>\n",
       "      <td>57.698052</td>\n",
       "      <td>10.04</td>\n",
       "      <td>5.74</td>\n",
       "      <td>5.94</td>\n",
       "      <td>6.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>FF3814A9-9F7D-E711-8530-00505688958B</td>\n",
       "      <td>0</td>\n",
       "      <td>2017-10-01</td>\n",
       "      <td>2100.0</td>\n",
       "      <td>345.0</td>\n",
       "      <td>158237.634409</td>\n",
       "      <td>22.58</td>\n",
       "      <td>10</td>\n",
       "      <td>78</td>\n",
       "      <td>15</td>\n",
       "      <td>...</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.93</td>\n",
       "      <td>8</td>\n",
       "      <td>7640</td>\n",
       "      <td>57.698052</td>\n",
       "      <td>10.04</td>\n",
       "      <td>5.74</td>\n",
       "      <td>5.94</td>\n",
       "      <td>6.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>FF3814A9-9F7D-E711-8530-00505688958B</td>\n",
       "      <td>3</td>\n",
       "      <td>2017-10-01</td>\n",
       "      <td>9614.9</td>\n",
       "      <td>769.4</td>\n",
       "      <td>123250.000000</td>\n",
       "      <td>78.81</td>\n",
       "      <td>8</td>\n",
       "      <td>50</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.93</td>\n",
       "      <td>8</td>\n",
       "      <td>7640</td>\n",
       "      <td>57.698052</td>\n",
       "      <td>10.04</td>\n",
       "      <td>5.74</td>\n",
       "      <td>5.94</td>\n",
       "      <td>6.03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>FF3814A9-9F7D-E711-8530-00505688958B</td>\n",
       "      <td>1</td>\n",
       "      <td>2017-11-01</td>\n",
       "      <td>11947.2</td>\n",
       "      <td>526.7</td>\n",
       "      <td>139420.634921</td>\n",
       "      <td>37.93</td>\n",
       "      <td>9</td>\n",
       "      <td>140</td>\n",
       "      <td>175</td>\n",
       "      <td>...</td>\n",
       "      <td>0.28</td>\n",
       "      <td>0.58</td>\n",
       "      <td>0.93</td>\n",
       "      <td>8</td>\n",
       "      <td>7640</td>\n",
       "      <td>58.926552</td>\n",
       "      <td>9.87</td>\n",
       "      <td>5.17</td>\n",
       "      <td>7.07</td>\n",
       "      <td>6.20</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 55 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 bulk_id  spalen       date1  start_square  \\\n",
       "id                                                                           \n",
       "0                                    NaN       1  2017-10-01       13109.9   \n",
       "1   FF3814A9-9F7D-E711-8530-00505688958B       2  2017-10-01       19898.4   \n",
       "2   FF3814A9-9F7D-E711-8530-00505688958B       0  2017-10-01        2100.0   \n",
       "3   FF3814A9-9F7D-E711-8530-00505688958B       3  2017-10-01        9614.9   \n",
       "4   FF3814A9-9F7D-E711-8530-00505688958B       1  2017-11-01       11947.2   \n",
       "\n",
       "     value          price  mean_sq  mean_fl  plan_s  plan_m  \\\n",
       "id                                                            \n",
       "0   1124.5  135633.913043      NaN        8     147     198   \n",
       "1   1412.3  128492.419825    58.01        9     139     120   \n",
       "2    345.0  158237.634409    22.58       10      78      15   \n",
       "3    769.4  123250.000000    78.81        8      50       0   \n",
       "4    526.7  139420.634921    37.93        9     140     175   \n",
       "\n",
       "           ...          До промки(км)  До парка(км)  До парка пешком(км)  \\\n",
       "id         ...                                                             \n",
       "0          ...                   0.28          0.58                 0.93   \n",
       "1          ...                   0.28          0.58                 0.93   \n",
       "2          ...                   0.28          0.58                 0.93   \n",
       "3          ...                   0.28          0.58                 0.93   \n",
       "4          ...                   0.28          0.58                 0.93   \n",
       "\n",
       "    Станций метро от кольца  Площадь двора       Курс Cтавка по ипотеке  \\\n",
       "id                                                                        \n",
       "0                         8           7640  57.698052             10.04   \n",
       "1                         8           7640  57.698052             10.04   \n",
       "2                         8           7640  57.698052             10.04   \n",
       "3                         8           7640  57.698052             10.04   \n",
       "4                         8           7640  58.926552              9.87   \n",
       "\n",
       "    Вклады до 1 года Вклады от 1 года до 3 лет  Вклады свыше 3 лет  \n",
       "id                                                                  \n",
       "0               5.74                      5.94                6.03  \n",
       "1               5.74                      5.94                6.03  \n",
       "2               5.74                      5.94                6.03  \n",
       "3               5.74                      5.94                6.03  \n",
       "4               5.17                      7.07                6.20  \n",
       "\n",
       "[5 rows x 55 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.loc[0::5, ['mean_sq','bulk_id']] = np.nan # каждая пятая строка содержит пропущенные\n",
    "df.head(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "все колонки, которые не являются числами, указываем категориальными (чтобы lgbm их съел):"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "for column in df.select_dtypes(include = ['object']).columns.tolist():\n",
    "    df[column] = df[column].astype('category')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "lgbm_regressor = lightgbm.LGBMRegressor().fit(df.loc[:, df.columns != 'value'], df['value'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "list_of_importances = list(zip(df.loc[:, df.columns != 'value'].columns.tolist(), \n",
    "                               lgbm_regressor.feature_importances_))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('bulk_id', 366),\n",
       " ('start_square', 339),\n",
       " ('date1', 285),\n",
       " ('plan_l', 173),\n",
       " ('mean_sq', 168),\n",
       " ('plan_m', 155),\n",
       " ('vid_2', 131),\n",
       " ('price', 129),\n",
       " ('plan_s', 116),\n",
       " ('vid_1', 115),\n",
       " ('Площадь двора', 98),\n",
       " ('spalen', 84),\n",
       " ('До парка(км)', 64),\n",
       " ('До большой дороги на машине(км)', 58),\n",
       " ('vid_0', 53),\n",
       " ('mean_fl', 51),\n",
       " ('До метро пешком(км)', 46),\n",
       " ('month_cnt', 45),\n",
       " ('Площадь зеленой зоны в радиусе 500 м', 44),\n",
       " ('month', 43),\n",
       " ('Вклады свыше 3 лет', 40),\n",
       " ('Количество помещений', 39),\n",
       " ('Вклады от 1 года до 3 лет', 34),\n",
       " ('До удобной авторазвязки на машине(км)', 32),\n",
       " ('До парка пешком(км)', 28),\n",
       " ('До промки(км)', 25),\n",
       " ('Площадь пром. зоны в радиусе 500 м', 23),\n",
       " ('Cтавка по ипотеке', 22),\n",
       " ('Курс', 21),\n",
       " ('Детский сад', 19),\n",
       " ('Школа', 19),\n",
       " ('Вклады до 1 года', 18),\n",
       " ('Площадь земельного участка', 16),\n",
       " ('Машиномест', 15),\n",
       " ('До Садового(км)', 14),\n",
       " ('Станций метро от кольца', 14),\n",
       " ('До Кремля', 12),\n",
       " ('До ТТК(км)', 11),\n",
       " ('Класс объекта', 9),\n",
       " ('Двор без машин', 9),\n",
       " ('ФОК', 5),\n",
       " ('Вентлияция', 5),\n",
       " ('Поликлиника', 3),\n",
       " ('Видеонаблюдение', 3),\n",
       " ('Кладовые', 1),\n",
       " ('Огорожена территория', 0),\n",
       " ('Входные группы', 0),\n",
       " ('Спортивная площадка', 0),\n",
       " ('Автомойка', 0),\n",
       " ('Колясочные', 0),\n",
       " ('Кондиционирование', 0),\n",
       " ('Лифт', 0),\n",
       " ('Система мусоротведения', 0),\n",
       " ('Подземная парковка', 0)]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sorted(list_of_importances, key= lambda x: x[1], reverse= True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['bulk_id',\n",
       " 'start_square',\n",
       " 'date1',\n",
       " 'plan_l',\n",
       " 'mean_sq',\n",
       " 'plan_m',\n",
       " 'vid_2',\n",
       " 'price',\n",
       " 'plan_s',\n",
       " 'vid_1',\n",
       " 'Площадь двора',\n",
       " 'spalen',\n",
       " 'До парка(км)',\n",
       " 'До большой дороги на машине(км)',\n",
       " 'vid_0',\n",
       " 'mean_fl',\n",
       " 'До метро пешком(км)',\n",
       " 'month_cnt',\n",
       " 'Площадь зеленой зоны в радиусе 500 м']"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "important_features = [x[0] for x in sorted(list_of_importances, \n",
    "                                           key= lambda x: x[1], \n",
    "                                           reverse= True)[:19]]\n",
    "important_features"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "оставляем в датасете только важные фичи:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "important_features.extend(['value'])\n",
    "df = df[important_features]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Лаба 3: пропуски, кодирование, масштабирование"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "заполнение пропусков:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "for column in df.select_dtypes(include= ['int64', 'float64']).columns.tolist():\n",
    "    df[column] = df[column].fillna(df[column].mean())\n",
    "for column in df.select_dtypes(include= ['category']).columns.tolist():\n",
    "    df[column] = df[column].fillna(df[column].describe(include= ['category'])['top'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "кодирование категориальных:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "for column in df.select_dtypes(include= ['category']).columns.tolist():\n",
    "    df = df.merge(df.groupby(column)['value'].mean().rename(str(column+'_value_mean')).to_frame(),\n",
    "                  how= 'left',\n",
    "                  left_on= column,\n",
    "                  right_index= True)\n",
    "    df.drop(column, axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "масштабирование признаков:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[:, df.columns != 'value'] = df.loc[:, df.columns != 'value'].apply(lambda x: x/x.max(), \n",
    "                                                                          axis=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Лаба 4: train-test, кросс-валидация, подбор параметров"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "трейн-тест:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "X_train, X_test, y_train, y_test = train_test_split(df.loc[:, df.columns != 'value'], \n",
    "                                                    df['value'], \n",
    "                                                    test_size= 0.1, \n",
    "                                                    random_state= 42)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "подбор лучшего параметра по кросс-валидации:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GridSearchCV(cv=3, error_score='raise-deprecating',\n",
       "       estimator=KNeighborsRegressor(algorithm='auto', leaf_size=30, metric='minkowski',\n",
       "          metric_params=None, n_jobs=None, n_neighbors=5, p=2,\n",
       "          weights='uniform'),\n",
       "       fit_params=None, iid='warn', n_jobs=None,\n",
       "       param_grid={'n_neighbors': [5, 10, 20, 50]},\n",
       "       pre_dispatch='2*n_jobs', refit=True, return_train_score='warn',\n",
       "       scoring='neg_mean_absolute_error', verbose=0)"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "random_search = GridSearchCV(estimator= KNeighborsRegressor(),\n",
    "                             param_grid= {'n_neighbors': [5,10,20,50]},\n",
    "                             scoring= 'neg_mean_absolute_error', \n",
    "                             cv= 3)\n",
    "\n",
    "random_search.fit(df.loc[:, df.columns != 'value'], df['value'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'n_neighbors': 50}"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "random_search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "202.2916783427088"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "-random_search.best_score_ # минус потому что метрика 'negative_mean_absolute_error'"
   ]
  }
 ],
 "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": "3.7.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}