deal_with_data_shifting.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# Create a method to find whether there is data shifting\n",
    "## And which features caused data shifting\n",
    "\n",
    "# Download the dataset here: https://www.kaggle.com/c/sberbank-russian-housing-market/data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from pandas import Series, DataFrame\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline\n",
    "\n",
    "from sklearn.ensemble import RandomForestRegressor\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn.cross_validation import cross_val_score\n",
    "from sklearn.preprocessing import LabelEncoder\n",
    "\n",
    "train = pd.read_csv('train.csv')\n",
    "test = pd.read_csv('test.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\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>timestamp</th>\n",
       "      <th>full_sq</th>\n",
       "      <th>life_sq</th>\n",
       "      <th>floor</th>\n",
       "      <th>max_floor</th>\n",
       "      <th>material</th>\n",
       "      <th>build_year</th>\n",
       "      <th>num_room</th>\n",
       "      <th>kitch_sq</th>\n",
       "      <th>...</th>\n",
       "      <th>cafe_count_5000_price_2500</th>\n",
       "      <th>cafe_count_5000_price_4000</th>\n",
       "      <th>cafe_count_5000_price_high</th>\n",
       "      <th>big_church_count_5000</th>\n",
       "      <th>church_count_5000</th>\n",
       "      <th>mosque_count_5000</th>\n",
       "      <th>leisure_count_5000</th>\n",
       "      <th>sport_count_5000</th>\n",
       "      <th>market_count_5000</th>\n",
       "      <th>price_doc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>2011-08-20</td>\n",
       "      <td>43</td>\n",
       "      <td>27.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>9</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>13</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>52</td>\n",
       "      <td>4</td>\n",
       "      <td>5850000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>2011-08-23</td>\n",
       "      <td>34</td>\n",
       "      <td>19.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>15</td>\n",
       "      <td>29</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>66</td>\n",
       "      <td>14</td>\n",
       "      <td>6000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>2011-08-27</td>\n",
       "      <td>43</td>\n",
       "      <td>29.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>27</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>67</td>\n",
       "      <td>10</td>\n",
       "      <td>5700000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>2011-09-01</td>\n",
       "      <td>89</td>\n",
       "      <td>50.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>26</td>\n",
       "      <td>3</td>\n",
       "      <td>13100000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>2011-09-05</td>\n",
       "      <td>77</td>\n",
       "      <td>77.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>319</td>\n",
       "      <td>108</td>\n",
       "      <td>17</td>\n",
       "      <td>135</td>\n",
       "      <td>236</td>\n",
       "      <td>2</td>\n",
       "      <td>91</td>\n",
       "      <td>195</td>\n",
       "      <td>14</td>\n",
       "      <td>16331452</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 292 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   id   timestamp  full_sq  life_sq  floor  max_floor  material  build_year  \\\n",
       "0   1  2011-08-20       43     27.0    4.0        NaN       NaN         NaN   \n",
       "1   2  2011-08-23       34     19.0    3.0        NaN       NaN         NaN   \n",
       "2   3  2011-08-27       43     29.0    2.0        NaN       NaN         NaN   \n",
       "3   4  2011-09-01       89     50.0    9.0        NaN       NaN         NaN   \n",
       "4   5  2011-09-05       77     77.0    4.0        NaN       NaN         NaN   \n",
       "\n",
       "   num_room  kitch_sq    ...      cafe_count_5000_price_2500  \\\n",
       "0       NaN       NaN    ...                               9   \n",
       "1       NaN       NaN    ...                              15   \n",
       "2       NaN       NaN    ...                              10   \n",
       "3       NaN       NaN    ...                              11   \n",
       "4       NaN       NaN    ...                             319   \n",
       "\n",
       "  cafe_count_5000_price_4000 cafe_count_5000_price_high  \\\n",
       "0                          4                          0   \n",
       "1                          3                          0   \n",
       "2                          3                          0   \n",
       "3                          2                          1   \n",
       "4                        108                         17   \n",
       "\n",
       "   big_church_count_5000  church_count_5000  mosque_count_5000  \\\n",
       "0                     13                 22                  1   \n",
       "1                     15                 29                  1   \n",
       "2                     11                 27                  0   \n",
       "3                      4                  4                  0   \n",
       "4                    135                236                  2   \n",
       "\n",
       "   leisure_count_5000  sport_count_5000  market_count_5000  price_doc  \n",
       "0                   0                52                  4    5850000  \n",
       "1                  10                66                 14    6000000  \n",
       "2                   4                67                 10    5700000  \n",
       "3                   0                26                  3   13100000  \n",
       "4                  91               195                 14   16331452  \n",
       "\n",
       "[5 rows x 292 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.head()\n",
    "\n",
    "# price_doc is the label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "id                                         int64\n",
       "timestamp                                 object\n",
       "full_sq                                    int64\n",
       "life_sq                                  float64\n",
       "floor                                    float64\n",
       "max_floor                                float64\n",
       "material                                 float64\n",
       "build_year                               float64\n",
       "num_room                                 float64\n",
       "kitch_sq                                 float64\n",
       "state                                    float64\n",
       "product_type                              object\n",
       "sub_area                                  object\n",
       "area_m                                   float64\n",
       "raion_popul                                int64\n",
       "green_zone_part                          float64\n",
       "indust_part                              float64\n",
       "children_preschool                         int64\n",
       "preschool_quota                          float64\n",
       "preschool_education_centers_raion          int64\n",
       "children_school                            int64\n",
       "school_quota                             float64\n",
       "school_education_centers_raion             int64\n",
       "school_education_centers_top_20_raion      int64\n",
       "hospital_beds_raion                      float64\n",
       "healthcare_centers_raion                   int64\n",
       "university_top_20_raion                    int64\n",
       "sport_objects_raion                        int64\n",
       "additional_education_raion                 int64\n",
       "culture_objects_top_25                    object\n",
       "                                          ...   \n",
       "big_church_count_3000                      int64\n",
       "church_count_3000                          int64\n",
       "mosque_count_3000                          int64\n",
       "leisure_count_3000                         int64\n",
       "sport_count_3000                           int64\n",
       "market_count_3000                          int64\n",
       "green_part_5000                          float64\n",
       "prom_part_5000                           float64\n",
       "office_count_5000                          int64\n",
       "office_sqm_5000                            int64\n",
       "trc_count_5000                             int64\n",
       "trc_sqm_5000                               int64\n",
       "cafe_count_5000                            int64\n",
       "cafe_sum_5000_min_price_avg              float64\n",
       "cafe_sum_5000_max_price_avg              float64\n",
       "cafe_avg_price_5000                      float64\n",
       "cafe_count_5000_na_price                   int64\n",
       "cafe_count_5000_price_500                  int64\n",
       "cafe_count_5000_price_1000                 int64\n",
       "cafe_count_5000_price_1500                 int64\n",
       "cafe_count_5000_price_2500                 int64\n",
       "cafe_count_5000_price_4000                 int64\n",
       "cafe_count_5000_price_high                 int64\n",
       "big_church_count_5000                      int64\n",
       "church_count_5000                          int64\n",
       "mosque_count_5000                          int64\n",
       "leisure_count_5000                         int64\n",
       "sport_count_5000                           int64\n",
       "market_count_5000                          int64\n",
       "price_doc                                  int64\n",
       "Length: 292, dtype: object"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\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>timestamp</th>\n",
       "      <th>full_sq</th>\n",
       "      <th>life_sq</th>\n",
       "      <th>floor</th>\n",
       "      <th>max_floor</th>\n",
       "      <th>material</th>\n",
       "      <th>build_year</th>\n",
       "      <th>num_room</th>\n",
       "      <th>kitch_sq</th>\n",
       "      <th>...</th>\n",
       "      <th>cafe_count_5000_price_1500</th>\n",
       "      <th>cafe_count_5000_price_2500</th>\n",
       "      <th>cafe_count_5000_price_4000</th>\n",
       "      <th>cafe_count_5000_price_high</th>\n",
       "      <th>big_church_count_5000</th>\n",
       "      <th>church_count_5000</th>\n",
       "      <th>mosque_count_5000</th>\n",
       "      <th>leisure_count_5000</th>\n",
       "      <th>sport_count_5000</th>\n",
       "      <th>market_count_5000</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>30474</td>\n",
       "      <td>2015-07-01</td>\n",
       "      <td>39.0</td>\n",
       "      <td>20.7</td>\n",
       "      <td>2</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>1998.0</td>\n",
       "      <td>1</td>\n",
       "      <td>8.9</td>\n",
       "      <td>...</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>14</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>30475</td>\n",
       "      <td>2015-07-01</td>\n",
       "      <td>79.2</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8</td>\n",
       "      <td>17</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>11</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>12</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>30476</td>\n",
       "      <td>2015-07-01</td>\n",
       "      <td>40.5</td>\n",
       "      <td>25.1</td>\n",
       "      <td>3</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>1960.0</td>\n",
       "      <td>2</td>\n",
       "      <td>4.8</td>\n",
       "      <td>...</td>\n",
       "      <td>42</td>\n",
       "      <td>11</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>71</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>30477</td>\n",
       "      <td>2015-07-01</td>\n",
       "      <td>62.8</td>\n",
       "      <td>36.0</td>\n",
       "      <td>17</td>\n",
       "      <td>17</td>\n",
       "      <td>1</td>\n",
       "      <td>2016.0</td>\n",
       "      <td>2</td>\n",
       "      <td>62.8</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>30478</td>\n",
       "      <td>2015-07-01</td>\n",
       "      <td>40.0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>17</td>\n",
       "      <td>17</td>\n",
       "      <td>1</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 291 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      id   timestamp  full_sq  life_sq  floor  max_floor  material  \\\n",
       "0  30474  2015-07-01     39.0     20.7      2          9         1   \n",
       "1  30475  2015-07-01     79.2      NaN      8         17         1   \n",
       "2  30476  2015-07-01     40.5     25.1      3          5         2   \n",
       "3  30477  2015-07-01     62.8     36.0     17         17         1   \n",
       "4  30478  2015-07-01     40.0     40.0     17         17         1   \n",
       "\n",
       "   build_year  num_room  kitch_sq        ...          \\\n",
       "0      1998.0         1       8.9        ...           \n",
       "1         0.0         3       1.0        ...           \n",
       "2      1960.0         2       4.8        ...           \n",
       "3      2016.0         2      62.8        ...           \n",
       "4         0.0         1       1.0        ...           \n",
       "\n",
       "   cafe_count_5000_price_1500 cafe_count_5000_price_2500  \\\n",
       "0                           8                          0   \n",
       "1                           4                          1   \n",
       "2                          42                         11   \n",
       "3                           1                          1   \n",
       "4                           5                          1   \n",
       "\n",
       "  cafe_count_5000_price_4000  cafe_count_5000_price_high  \\\n",
       "0                          0                           0   \n",
       "1                          1                           0   \n",
       "2                          4                           0   \n",
       "3                          2                           0   \n",
       "4                          1                           0   \n",
       "\n",
       "   big_church_count_5000  church_count_5000  mosque_count_5000  \\\n",
       "0                      1                 10                  1   \n",
       "1                      2                 11                  0   \n",
       "2                     10                 21                  0   \n",
       "3                      0                 10                  0   \n",
       "4                      2                 12                  0   \n",
       "\n",
       "   leisure_count_5000  sport_count_5000  market_count_5000  \n",
       "0                   0                14                  1  \n",
       "1                   1                12                  1  \n",
       "2                  10                71                 11  \n",
       "3                   0                 2                  0  \n",
       "4                   1                11                  1  \n",
       "\n",
       "[5 rows x 291 columns]"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Preprocess data first, before checking data shifting\n",
    "\n",
    "## Sometimes, the lowercase, uppercase for categorical data matters, here skip checking data consistency\n",
    "\n",
    "\n",
    "## Deal with missing Data\n",
    "for i in train.columns:\n",
    "    if train[i].dtype == 'object':\n",
    "        train[i] = train[i].fillna(train[i].mode().iloc[0])\n",
    "    if (train[i].dtype == 'int64' or train[i].dtype == 'float64'):\n",
    "        train[i] = train[i].fillna(np.nanmedian(train[i]))\n",
    "    \n",
    "for i in test.columns:\n",
    "    if test[i].dtype == 'object':\n",
    "        test[i] = test[i].fillna(test[i].mode().iloc[0])\n",
    "    if (test[i].dtype == 'int64' or test[i].dtype == 'float64'):\n",
    "        test[i] = test[i].fillna(np.nanmedian(test[i]))    \n",
    "# with np.nanmedian, you can get median by ignoring NaN, otherwise, statistics.median could use NaN as median"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\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>timestamp</th>\n",
       "      <th>full_sq</th>\n",
       "      <th>life_sq</th>\n",
       "      <th>floor</th>\n",
       "      <th>max_floor</th>\n",
       "      <th>material</th>\n",
       "      <th>build_year</th>\n",
       "      <th>num_room</th>\n",
       "      <th>kitch_sq</th>\n",
       "      <th>...</th>\n",
       "      <th>cafe_count_5000_price_2500</th>\n",
       "      <th>cafe_count_5000_price_4000</th>\n",
       "      <th>cafe_count_5000_price_high</th>\n",
       "      <th>big_church_count_5000</th>\n",
       "      <th>church_count_5000</th>\n",
       "      <th>mosque_count_5000</th>\n",
       "      <th>leisure_count_5000</th>\n",
       "      <th>sport_count_5000</th>\n",
       "      <th>market_count_5000</th>\n",
       "      <th>price_doc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>2011-08-20</td>\n",
       "      <td>43</td>\n",
       "      <td>27.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1979.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>9</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>13</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>52</td>\n",
       "      <td>4</td>\n",
       "      <td>5850000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>2011-08-23</td>\n",
       "      <td>34</td>\n",
       "      <td>19.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1979.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>15</td>\n",
       "      <td>29</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>66</td>\n",
       "      <td>14</td>\n",
       "      <td>6000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>2011-08-27</td>\n",
       "      <td>43</td>\n",
       "      <td>29.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1979.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>27</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>67</td>\n",
       "      <td>10</td>\n",
       "      <td>5700000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>2011-09-01</td>\n",
       "      <td>89</td>\n",
       "      <td>50.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1979.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>26</td>\n",
       "      <td>3</td>\n",
       "      <td>13100000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>2011-09-05</td>\n",
       "      <td>77</td>\n",
       "      <td>77.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1979.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>319</td>\n",
       "      <td>108</td>\n",
       "      <td>17</td>\n",
       "      <td>135</td>\n",
       "      <td>236</td>\n",
       "      <td>2</td>\n",
       "      <td>91</td>\n",
       "      <td>195</td>\n",
       "      <td>14</td>\n",
       "      <td>16331452</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 292 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   id   timestamp  full_sq  life_sq  floor  max_floor  material  build_year  \\\n",
       "0   1  2011-08-20       43     27.0    4.0       12.0       1.0      1979.0   \n",
       "1   2  2011-08-23       34     19.0    3.0       12.0       1.0      1979.0   \n",
       "2   3  2011-08-27       43     29.0    2.0       12.0       1.0      1979.0   \n",
       "3   4  2011-09-01       89     50.0    9.0       12.0       1.0      1979.0   \n",
       "4   5  2011-09-05       77     77.0    4.0       12.0       1.0      1979.0   \n",
       "\n",
       "   num_room  kitch_sq    ...      cafe_count_5000_price_2500  \\\n",
       "0       2.0       6.0    ...                               9   \n",
       "1       2.0       6.0    ...                              15   \n",
       "2       2.0       6.0    ...                              10   \n",
       "3       2.0       6.0    ...                              11   \n",
       "4       2.0       6.0    ...                             319   \n",
       "\n",
       "  cafe_count_5000_price_4000 cafe_count_5000_price_high  \\\n",
       "0                          4                          0   \n",
       "1                          3                          0   \n",
       "2                          3                          0   \n",
       "3                          2                          1   \n",
       "4                        108                         17   \n",
       "\n",
       "   big_church_count_5000  church_count_5000  mosque_count_5000  \\\n",
       "0                     13                 22                  1   \n",
       "1                     15                 29                  1   \n",
       "2                     11                 27                  0   \n",
       "3                      4                  4                  0   \n",
       "4                    135                236                  2   \n",
       "\n",
       "   leisure_count_5000  sport_count_5000  market_count_5000  price_doc  \n",
       "0                   0                52                  4    5850000  \n",
       "1                  10                66                 14    6000000  \n",
       "2                   4                67                 10    5700000  \n",
       "3                   0                26                  3   13100000  \n",
       "4                  91               195                 14   16331452  \n",
       "\n",
       "[5 rows x 292 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['2011-08-20', '2011-08-23', '2011-08-27', ..., '2015-06-27',\n",
       "       '2015-06-29', '2015-06-30'], dtype=object)"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.timestamp.unique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "## Categorical to Numerical (people call it label encoding, I think this name is confusing)\n",
    "number = LabelEncoder()\n",
    "for i in train.columns:\n",
    "    if (train[i].dtype == 'object'):\n",
    "        train[i] = number.fit_transform(train[i].astype('str'))\n",
    "        train[i] = train[i].astype('object')\n",
    "\n",
    "for i in test.columns:\n",
    "    if (test[i].dtype == 'object'):\n",
    "        test[i] = number.fit_transform(test[i].astype('str'))\n",
    "        test[i] = test[i].astype('object')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\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>timestamp</th>\n",
       "      <th>full_sq</th>\n",
       "      <th>life_sq</th>\n",
       "      <th>floor</th>\n",
       "      <th>max_floor</th>\n",
       "      <th>material</th>\n",
       "      <th>build_year</th>\n",
       "      <th>num_room</th>\n",
       "      <th>kitch_sq</th>\n",
       "      <th>...</th>\n",
       "      <th>cafe_count_5000_price_2500</th>\n",
       "      <th>cafe_count_5000_price_4000</th>\n",
       "      <th>cafe_count_5000_price_high</th>\n",
       "      <th>big_church_count_5000</th>\n",
       "      <th>church_count_5000</th>\n",
       "      <th>mosque_count_5000</th>\n",
       "      <th>leisure_count_5000</th>\n",
       "      <th>sport_count_5000</th>\n",
       "      <th>market_count_5000</th>\n",
       "      <th>price_doc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>43</td>\n",
       "      <td>27.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1979.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>9</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>13</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>52</td>\n",
       "      <td>4</td>\n",
       "      <td>5850000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>34</td>\n",
       "      <td>19.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1979.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>15</td>\n",
       "      <td>29</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>66</td>\n",
       "      <td>14</td>\n",
       "      <td>6000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>43</td>\n",
       "      <td>29.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1979.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>27</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>67</td>\n",
       "      <td>10</td>\n",
       "      <td>5700000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>89</td>\n",
       "      <td>50.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1979.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>26</td>\n",
       "      <td>3</td>\n",
       "      <td>13100000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>4</td>\n",
       "      <td>77</td>\n",
       "      <td>77.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1979.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>...</td>\n",
       "      <td>319</td>\n",
       "      <td>108</td>\n",
       "      <td>17</td>\n",
       "      <td>135</td>\n",
       "      <td>236</td>\n",
       "      <td>2</td>\n",
       "      <td>91</td>\n",
       "      <td>195</td>\n",
       "      <td>14</td>\n",
       "      <td>16331452</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 292 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   id timestamp  full_sq  life_sq  floor  max_floor  material  build_year  \\\n",
       "0   1         0       43     27.0    4.0       12.0       1.0      1979.0   \n",
       "1   2         1       34     19.0    3.0       12.0       1.0      1979.0   \n",
       "2   3         2       43     29.0    2.0       12.0       1.0      1979.0   \n",
       "3   4         3       89     50.0    9.0       12.0       1.0      1979.0   \n",
       "4   5         4       77     77.0    4.0       12.0       1.0      1979.0   \n",
       "\n",
       "   num_room  kitch_sq    ...      cafe_count_5000_price_2500  \\\n",
       "0       2.0       6.0    ...                               9   \n",
       "1       2.0       6.0    ...                              15   \n",
       "2       2.0       6.0    ...                              10   \n",
       "3       2.0       6.0    ...                              11   \n",
       "4       2.0       6.0    ...                             319   \n",
       "\n",
       "  cafe_count_5000_price_4000 cafe_count_5000_price_high  \\\n",
       "0                          4                          0   \n",
       "1                          3                          0   \n",
       "2                          3                          0   \n",
       "3                          2                          1   \n",
       "4                        108                         17   \n",
       "\n",
       "   big_church_count_5000  church_count_5000  mosque_count_5000  \\\n",
       "0                     13                 22                  1   \n",
       "1                     15                 29                  1   \n",
       "2                     11                 27                  0   \n",
       "3                      4                  4                  0   \n",
       "4                    135                236                  2   \n",
       "\n",
       "   leisure_count_5000  sport_count_5000  market_count_5000  price_doc  \n",
       "0                   0                52                  4    5850000  \n",
       "1                  10                66                 14    6000000  \n",
       "2                   4                67                 10    5700000  \n",
       "3                   0                26                  3   13100000  \n",
       "4                  91               195                 14   16331452  \n",
       "\n",
       "[5 rows x 292 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "## create this new col \"origin\" to tell which set comes from train, which comes from test\n",
    "train['origin'] = 0\n",
    "test['origin'] = 1\n",
    "training = train.drop('price_doc',axis=1) # drop the label"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "## taking sample from training and test data\n",
    "training = training.sample(9000, random_state=77)\n",
    "testing = test.sample(7000, random_state=99)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "## combining random samples, seperate 'origin' col with the data\n",
    "combi = training.append(testing)\n",
    "y = combi['origin']\n",
    "combi.drop('origin',axis=1,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('id', 1.0)\n",
      "('timestamp', 0.95935706349206351)\n",
      "('kitch_sq', 0.89323138888888887)\n",
      "('cafe_sum_500_min_price_avg', 0.83561873015873012)\n",
      "('cafe_avg_price_500', 0.83495988095238105)\n"
     ]
    }
   ],
   "source": [
    "# Find drifting features can be dropped, put into drop_list\n",
    "## If a feature has average score > 0.8, put into drop_list. \n",
    "## Because it means the feature could decide it's \"train\" or \"test\"\n",
    "\n",
    "model = RandomForestClassifier(n_estimators = 50, max_depth = 5, min_samples_leaf = 5)\n",
    "drop_list = []\n",
    "\n",
    "for i in combi.columns:\n",
    "    score = cross_val_score(model,pd.DataFrame(combi[i]),y,cv=5,scoring='roc_auc')\n",
    "    if (np.mean(score) > 0.8):\n",
    "        drop_list.append(i)\n",
    "        print(i,np.mean(score))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "# How to treat drifting features in order to improve final model prediction\n",
    "\n",
    "training = train.drop('origin',axis=1)\n",
    "testing = test.drop('origin',axis=1)\n",
    "\n",
    "# using regressor here is because of the label is numerical\n",
    "rf = RandomForestRegressor(n_estimators=200, max_depth=4,max_features=10)\n",
    "\n",
    "rf.fit(training.drop('price_doc',axis=1),training['price_doc'])   # it ignores id automatically\n",
    "pred = rf.predict(testing)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbUAAAKSCAYAAABV60XOAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x111ac2050>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Dropping drifting features but check feature importance first\n",
    "## plotting feature importance (top 10) and find whether there is drifting features are important\n",
    "features = training.columns.values\n",
    "imp = rf.feature_importances_\n",
    "indices = np.argsort(imp)[::-1][:10]\n",
    "\n",
    "#plot\n",
    "plt.figure(figsize=(7,9))\n",
    "plt.bar(range(len(indices)), imp[indices], color = 'purple', align='center')\n",
    "plt.xticks(range(len(indices)), features[indices], rotation='vertical')\n",
    "plt.xlim([-1,len(indices)])\n",
    "plt.show()\n",
    "\n",
    "# kitch_sq is important, although it's showing drifting, keep this feature"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestRegressor(bootstrap=True, criterion='mse', max_depth=4,\n",
       "           max_features=10, max_leaf_nodes=None, min_samples_leaf=1,\n",
       "           min_samples_split=2, min_weight_fraction_leaf=0.0,\n",
       "           n_estimators=200, n_jobs=1, oob_score=False, random_state=None,\n",
       "           verbose=0, warm_start=False)"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## dropping drifting features which are not important.\n",
    "drift_train = training.drop(['id','cafe_sum_500_min_price_avg','cafe_avg_price_500'], axis=1)\n",
    "drift_test = testing.drop(['id','cafe_sum_500_min_price_avg','cafe_avg_price_500'], axis=1)\n",
    "\n",
    "rf = RandomForestRegressor(n_estimators=200, max_depth=4,max_features=10)\n",
    "rf.fit(drift_train.drop('price_doc',axis=1),training['price_doc'])\n",
    "\n",
    "# submit your result to Kaggle: https://www.kaggle.com/c/sberbank-russian-housing-market/data\n",
    "pred = rf.predict(drift_test)\n",
    "columns = ['price_doc']\n",
    "sub = pd.DataFrame(data=pred,columns=columns)\n",
    "sub['id'] = test['id']\n",
    "sub = sub[['id','price_doc']]\n",
    "sub.to_csv('without_drifting.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# method 2 - MLBox\n",
    "\n",
    "from mlbox.preprocessing import *"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\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>timestamp</th>\n",
       "      <th>full_sq</th>\n",
       "      <th>life_sq</th>\n",
       "      <th>floor</th>\n",
       "      <th>max_floor</th>\n",
       "      <th>material</th>\n",
       "      <th>build_year</th>\n",
       "      <th>num_room</th>\n",
       "      <th>kitch_sq</th>\n",
       "      <th>...</th>\n",
       "      <th>cafe_count_5000_price_2500</th>\n",
       "      <th>cafe_count_5000_price_4000</th>\n",
       "      <th>cafe_count_5000_price_high</th>\n",
       "      <th>big_church_count_5000</th>\n",
       "      <th>church_count_5000</th>\n",
       "      <th>mosque_count_5000</th>\n",
       "      <th>leisure_count_5000</th>\n",
       "      <th>sport_count_5000</th>\n",
       "      <th>market_count_5000</th>\n",
       "      <th>price_doc</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>2011-08-20</td>\n",
       "      <td>43</td>\n",
       "      <td>27.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>9</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>13</td>\n",
       "      <td>22</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>52</td>\n",
       "      <td>4</td>\n",
       "      <td>5850000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2</td>\n",
       "      <td>2011-08-23</td>\n",
       "      <td>34</td>\n",
       "      <td>19.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>15</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>15</td>\n",
       "      <td>29</td>\n",
       "      <td>1</td>\n",
       "      <td>10</td>\n",
       "      <td>66</td>\n",
       "      <td>14</td>\n",
       "      <td>6000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>3</td>\n",
       "      <td>2011-08-27</td>\n",
       "      <td>43</td>\n",
       "      <td>29.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>10</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>11</td>\n",
       "      <td>27</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>67</td>\n",
       "      <td>10</td>\n",
       "      <td>5700000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>2011-09-01</td>\n",
       "      <td>89</td>\n",
       "      <td>50.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>11</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>26</td>\n",
       "      <td>3</td>\n",
       "      <td>13100000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>5</td>\n",
       "      <td>2011-09-05</td>\n",
       "      <td>77</td>\n",
       "      <td>77.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>...</td>\n",
       "      <td>319</td>\n",
       "      <td>108</td>\n",
       "      <td>17</td>\n",
       "      <td>135</td>\n",
       "      <td>236</td>\n",
       "      <td>2</td>\n",
       "      <td>91</td>\n",
       "      <td>195</td>\n",
       "      <td>14</td>\n",
       "      <td>16331452</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 292 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "   id   timestamp  full_sq  life_sq  floor  max_floor  material  build_year  \\\n",
       "0   1  2011-08-20       43     27.0    4.0        NaN       NaN         NaN   \n",
       "1   2  2011-08-23       34     19.0    3.0        NaN       NaN         NaN   \n",
       "2   3  2011-08-27       43     29.0    2.0        NaN       NaN         NaN   \n",
       "3   4  2011-09-01       89     50.0    9.0        NaN       NaN         NaN   \n",
       "4   5  2011-09-05       77     77.0    4.0        NaN       NaN         NaN   \n",
       "\n",
       "   num_room  kitch_sq    ...      cafe_count_5000_price_2500  \\\n",
       "0       NaN       NaN    ...                               9   \n",
       "1       NaN       NaN    ...                              15   \n",
       "2       NaN       NaN    ...                              10   \n",
       "3       NaN       NaN    ...                              11   \n",
       "4       NaN       NaN    ...                             319   \n",
       "\n",
       "  cafe_count_5000_price_4000 cafe_count_5000_price_high  \\\n",
       "0                          4                          0   \n",
       "1                          3                          0   \n",
       "2                          3                          0   \n",
       "3                          2                          1   \n",
       "4                        108                         17   \n",
       "\n",
       "   big_church_count_5000  church_count_5000  mosque_count_5000  \\\n",
       "0                     13                 22                  1   \n",
       "1                     15                 29                  1   \n",
       "2                     11                 27                  0   \n",
       "3                      4                  4                  0   \n",
       "4                    135                236                  2   \n",
       "\n",
       "   leisure_count_5000  sport_count_5000  market_count_5000  price_doc  \n",
       "0                   0                52                  4    5850000  \n",
       "1                  10                66                 14    6000000  \n",
       "2                   4                67                 10    5700000  \n",
       "3                   0                26                  3   13100000  \n",
       "4                  91               195                 14   16331452  \n",
       "\n",
       "[5 rows x 292 columns]"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train.head()   # without cleaning, mlbox does cleanning for you"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['id', 'timestamp', 'full_sq', 'life_sq', 'floor', 'max_floor',\n",
       "       'material', 'build_year', 'num_room', 'kitch_sq',\n",
       "       ...\n",
       "       'cafe_count_5000_price_2500', 'cafe_count_5000_price_4000',\n",
       "       'cafe_count_5000_price_high', 'big_church_count_5000',\n",
       "       'church_count_5000', 'mosque_count_5000', 'leisure_count_5000',\n",
       "       'sport_count_5000', 'market_count_5000', 'price_doc'],\n",
       "      dtype='object', length=292)"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cols = train.columns\n",
    "cols"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "reading csv : train.csv ...\n",
      "cleaning data...\n",
      "CPU time: 38.210623025894165 seconds\n",
      "\n",
      "reading csv : test.csv ...\n",
      "cleaning data...\n",
      "CPU time: 1.2265539169311523 seconds\n",
      "\n",
      "number of common features : 294\n",
      "\n",
      "gathering and crunching for train and test datasets\n",
      "reindexing for train and test datasets\n",
      "dropping training duplicates\n",
      "dropping constant variables on training set\n",
      "\n",
      "number of categorical features : 15\n",
      "number of numerical features : 279\n",
      "number of training samples : 30471\n",
      "number of test samples : 7662\n",
      "\n",
      "Top sparse features (% missing values on train set):\n",
      "hospital_beds_raion           47.4\n",
      "build_year                    44.6\n",
      "state                         44.5\n",
      "cafe_avg_price_500            43.6\n",
      "cafe_sum_500_max_price_avg    43.6\n",
      "dtype: float64\n",
      "\n",
      "task : regression\n"
     ]
    }
   ],
   "source": [
    "data = Reader(sep=\",\").train_test_split([\"train.csv\", \"test.csv\"], \"price_doc\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "computing drifts...\n",
      "CPU time: 4.431854009628296 seconds\n",
      "\n",
      "Top 10 drifts\n",
      "\n",
      "('id', 1.0)\n",
      "('timestamp_TIMESTAMP', 1.0)\n",
      "('kitch_sq', 0.85066615773743659)\n",
      "('state', 0.69643064512586084)\n",
      "('life_sq', 0.68690434781272502)\n",
      "('material', 0.67251970247347859)\n",
      "('build_year', 0.6713825597594798)\n",
      "('max_floor', 0.66605720777439148)\n",
      "('num_room', 0.66429234043615915)\n",
      "('full_sq', 0.65005102278248572)\n",
      "\n",
      "deleted variables : ['id', 'kitch_sq', 'timestamp_TIMESTAMP']\n",
      "\n",
      "dumping drift coefficients into directory : save\n",
      "drift coefficients dumped\n"
     ]
    }
   ],
   "source": [
    "#deleting non-stable/shifting features\n",
    "## 0.8 is the default threshold\n",
    "\n",
    "data = Drift_thresholder().fit_transform(data)  "
   ]
  }
 ],
 "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": "3.6.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}