Using feature tools-Copy1.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "toc": true
   },
   "source": [
    "<h1>Table of Contents<span class=\"tocSkip\"></span></h1>\n",
    "<div class=\"toc\"><ul class=\"toc-item\"><li><span><a href=\"#Feature-tools-on-entirety\" data-toc-modified-id=\"Feature-tools-on-entirety-1\"><span class=\"toc-item-num\">1&nbsp;&nbsp;</span>Feature tools on entirety</a></span></li><li><span><a href=\"#Interactions-in-test&amp;train-dfs\" data-toc-modified-id=\"Interactions-in-test&amp;train-dfs-2\"><span class=\"toc-item-num\">2&nbsp;&nbsp;</span>Interactions in test&amp;train dfs</a></span></li><li><span><a href=\"#Merge-automated-&amp;-manual-aggs\" data-toc-modified-id=\"Merge-automated-&amp;-manual-aggs-3\"><span class=\"toc-item-num\">3&nbsp;&nbsp;</span>Merge automated &amp; manual aggs</a></span></li></ul></div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/javascript": [
       "IPython.notebook.set_autosave_interval(120000)"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Autosaving every 120 seconds\n"
     ]
    }
   ],
   "source": [
    "%matplotlib inline\n",
    "%load_ext autoreload\n",
    "%autoreload 4\n",
    "%autosave 120"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from fastai.io import *\n",
    "from fastai.structured import *\n",
    "from sklearn.ensemble import RandomForestRegressor, RandomForestClassifier\n",
    "from pandas_summary import DataFrameSummary\n",
    "from IPython.display import display\n",
    "from sklearn import metrics\n",
    "import feather\n",
    "import gc\n",
    "import featuretools as ft"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gc.collect()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "temp_hist_trans = feather.read_dataframe('hist_trans_beta')\n",
    "temp_new_trans = feather.read_dataframe('new_hist_trans_beta')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "def additional_feats(hist_df):\n",
    "#     hist_df['price'] = hist_df['purchase_amount'] / hist_df['installments']\n",
    "    hist_df['duration'] = hist_df['purchase_amount']*hist_df['month_diff']\n",
    "    hist_df['amount_month_ratio'] = hist_df['purchase_amount']/hist_df['month_diff']\n",
    "    return hist_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "dfs = [temp_hist_trans, temp_new_trans]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "temp_hist_trans, temp_new_trans = [additional_feats(df) for df in dfs]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "for df in [temp_hist_trans, temp_new_trans]:\n",
    "    df.drop(['purchase_Year', 'purchase_Month', 'purchase_Week', 'purchase_Day',\n",
    "       'purchase_Dayofweek', 'purchase_Dayofyear', 'purchase_Is_month_end', 'purchase_Is_month_start',\n",
    "       'purchase_Is_quarter_end', 'purchase_Is_quarter_start', 'purchase_Is_year_end',\n",
    "       'purchase_Is_year_start', 'purchase_Elapsed'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((29112361, 20), (1963031, 20))"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp_hist_trans.shape, temp_new_trans.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(['authorized_flag', 'card_id', 'city_id', 'category_1', 'installments', 'category_3',\n",
       "       'merchant_category_id', 'merchant_id', 'month_lag', 'purchase_amount', 'purchase_date', 'category_2',\n",
       "       'state_id', 'subsector_id', 'purchased_on_weekend', 'purchased_on_weekday', 'month_diff',\n",
       "       'purchase_date_successive_diff', 'duration', 'amount_month_ratio'], dtype=object)"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp_new_trans.columns.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "cat_cols = ['authorized_flag', 'card_id', 'city_id', 'category_1', 'category_3',\n",
    "       'merchant_category_id', 'merchant_id', 'category_2',\n",
    "       'state_id', 'subsector_id', 'purchased_on_weekend']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "cat_dict = {c:ft.variable_types.Categorical for c in cat_cols}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "cat_dict['purchase_date'] = ft.variable_types.Datetime"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'authorized_flag': featuretools.variable_types.variable.Categorical,\n",
       " 'card_id': featuretools.variable_types.variable.Categorical,\n",
       " 'city_id': featuretools.variable_types.variable.Categorical,\n",
       " 'category_1': featuretools.variable_types.variable.Categorical,\n",
       " 'category_3': featuretools.variable_types.variable.Categorical,\n",
       " 'merchant_category_id': featuretools.variable_types.variable.Categorical,\n",
       " 'merchant_id': featuretools.variable_types.variable.Categorical,\n",
       " 'category_2': featuretools.variable_types.variable.Categorical,\n",
       " 'state_id': featuretools.variable_types.variable.Categorical,\n",
       " 'subsector_id': featuretools.variable_types.variable.Categorical,\n",
       " 'purchased_on_weekend': featuretools.variable_types.variable.Categorical,\n",
       " 'purchase_date': featuretools.variable_types.variable.Datetime}"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cat_dict"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature tools on entirety"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "68"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gc.collect()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "es = ft.EntitySet()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2019-02-07 00:58:01,439 featuretools.entityset - WARNING    index index not found in dataframe, creating new integer column\n"
     ]
    }
   ],
   "source": [
    "temp_es = es.entity_from_dataframe('temp_hist_trans', temp_hist_trans, 'index', time_index='purchase_date',\n",
    "                                        variable_types=cat_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "35"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gc.collect()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2019-02-07 00:58:55,186 featuretools.entityset - WARNING    index index not found in dataframe, creating new integer column\n"
     ]
    }
   ],
   "source": [
    "temp_es = es.entity_from_dataframe('temp_new_trans', temp_new_trans, 'index', time_index='purchase_date',\n",
    "                                        variable_types=cat_dict)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "35"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gc.collect()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "PATH = 'data/elo/'\n",
    "train, test = [pd.read_csv(f'{PATH}{c}') for c in ['train.csv', 'test.csv']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "temp_train = train"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "temp_test = test"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((201917, 6), (123623, 5))"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp_train.shape, temp_test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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>first_active_month</th>\n",
       "      <th>card_id</th>\n",
       "      <th>feature_1</th>\n",
       "      <th>feature_2</th>\n",
       "      <th>feature_3</th>\n",
       "      <th>target</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-06</td>\n",
       "      <td>C_ID_92a2005557</td>\n",
       "      <td>5</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>-0.820283</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-01</td>\n",
       "      <td>C_ID_3d0044924f</td>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0.392913</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2016-08</td>\n",
       "      <td>C_ID_d639edf6cd</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0.688056</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-09</td>\n",
       "      <td>C_ID_186d6a6901</td>\n",
       "      <td>4</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0.142495</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2017-11</td>\n",
       "      <td>C_ID_cdbd2c0db2</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>-0.159749</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  first_active_month          card_id  feature_1  feature_2  feature_3  \\\n",
       "0            2017-06  C_ID_92a2005557          5          2          1   \n",
       "1            2017-01  C_ID_3d0044924f          4          1          0   \n",
       "2            2016-08  C_ID_d639edf6cd          2          2          0   \n",
       "3            2017-09  C_ID_186d6a6901          4          3          0   \n",
       "4            2017-11  C_ID_cdbd2c0db2          1          3          0   \n",
       "\n",
       "     target  \n",
       "0 -0.820283  \n",
       "1  0.392913  \n",
       "2  0.688056  \n",
       "3  0.142495  \n",
       "4 -0.159749  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp_train.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "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>first_active_month</th>\n",
       "      <th>card_id</th>\n",
       "      <th>feature_1</th>\n",
       "      <th>feature_2</th>\n",
       "      <th>feature_3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-04</td>\n",
       "      <td>C_ID_0ab67a22ab</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-01</td>\n",
       "      <td>C_ID_130fd0cbdd</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-08</td>\n",
       "      <td>C_ID_b709037bc5</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-12</td>\n",
       "      <td>C_ID_d27d835a9f</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2015-12</td>\n",
       "      <td>C_ID_2b5e3df5c2</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  first_active_month          card_id  feature_1  feature_2  feature_3\n",
       "0            2017-04  C_ID_0ab67a22ab          3          3          1\n",
       "1            2017-01  C_ID_130fd0cbdd          2          3          0\n",
       "2            2017-08  C_ID_b709037bc5          5          1          1\n",
       "3            2017-12  C_ID_d27d835a9f          2          1          0\n",
       "4            2015-12  C_ID_2b5e3df5c2          5          1          1"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp_test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_cat_cols = ['feature_1', 'feature_2', 'feature_3']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_cat_cols = {c:ft.variable_types.Ordinal for c in train_cat_cols}"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "temp_es = es.entity_from_dataframe('train', temp_train, 'card_id', time_index='first_active_month',\n",
    "                                        variable_types=train_cat_cols)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Entity: train\n",
       "  Variables:\n",
       "    card_id (dtype: index)\n",
       "    first_active_month (dtype: datetime_time_index)\n",
       "    target (dtype: numeric)\n",
       "    feature_1 (dtype: ordinal)\n",
       "    feature_2 (dtype: ordinal)\n",
       "    feature_3 (dtype: ordinal)\n",
       "  Shape:\n",
       "    (Rows: 201917, Columns: 6)"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp_es['train']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "temp_es = es.entity_from_dataframe('test', temp_train, 'card_id', time_index='first_active_month',\n",
    "                                        variable_types=train_cat_cols)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "new_r1 = ft.Relationship( temp_es['train']['card_id'],temp_es['temp_hist_trans']['card_id'])\n",
    "new_r2 = ft.Relationship( temp_es['train']['card_id'],temp_es['temp_new_trans']['card_id'])\n",
    "new_r3 = ft.Relationship( temp_es['test']['card_id'],temp_es['temp_hist_trans']['card_id'])\n",
    "new_r4 = ft.Relationship( temp_es['test']['card_id'],temp_es['temp_new_trans']['card_id'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for r in [new_r1, new_r2, new_r3, new_r4]:\n",
    "    \n",
    "    temp_es = temp_es.add_relationship(r)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Entityset: None\n",
       "  Entities:\n",
       "    temp_hist_trans [Rows: 29112361, Columns: 21]\n",
       "    temp_new_trans [Rows: 1963031, Columns: 21]\n",
       "    train [Rows: 201917, Columns: 6]\n",
       "    test [Rows: 201917, Columns: 6]\n",
       "  Relationships:\n",
       "    temp_hist_trans.card_id -> train.card_id\n",
       "    temp_new_trans.card_id -> train.card_id\n",
       "    temp_hist_trans.card_id -> test.card_id\n",
       "    temp_new_trans.card_id -> test.card_id"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "temp_es"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "default_agg_primitives =  [\"sum\", \"std\", \"max\", \"skew\", \"min\", \"mean\", \"count\", \"median\",\n",
    "                           \"percent_true\", \"num_unique\", \"mode\", \"avg_time_between\", \"time_since_last\",\n",
    "                           \"percent_true\", \"trend\"]\n",
    "default_trans_primitives =  [\"day\", \"month\",\"weekday\",\"days_since\", \"percentile\", \"hour\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "temp_es['temp_hist_trans']['authorized_flag'].interesting_values = [1]\n",
    "# temp_es['temp_hist_trans']['purchased_on_weekend'].interesting_values = [1]\n",
    "# temp_es['temp_new_trans']['purchased_on_weekend'].interesting_values = [1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Built 533 features\n",
      "Elapsed: 8:09:44 | Remaining: 266:19:26 | Progress:   3%|▎         | Calculated: 6/202 chunks"
     ]
    }
   ],
   "source": [
    "%%time \n",
    "feature_matrix, feature_names = ft.dfs(entityset = temp_es,target_entity = 'train',\n",
    "                                             trans_primitives =default_trans_primitives,\n",
    "                                             agg_primitives=default_agg_primitives,\n",
    "                                             ignore_entities=['test'],\n",
    "                                             where_primitives = ['mean', 'sum'],\n",
    "                                             chunk_size=1000,\n",
    "                                             max_depth = 2,\n",
    "                                             features_only=False,\n",
    "                                             verbose = True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "gc.collect()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_matrix.to_csv('inter.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_matrix.reset_index().to_feather('feature_matrix')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "collapsed": true
   },
   "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>card_id</th>\n",
       "      <th>C_ID_92a2005557</th>\n",
       "      <th>C_ID_ae86f7d5fb</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>target</th>\n",
       "      <td>-0.820283</td>\n",
       "      <td>-0.563362</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>feature_1</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>feature_2</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>feature_3</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.installments)</th>\n",
       "      <td>4</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.month_lag)</th>\n",
       "      <td>-1017</td>\n",
       "      <td>-308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.purchase_amount)</th>\n",
       "      <td>-165.969</td>\n",
       "      <td>-74.8775</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.purchased_on_weekday)</th>\n",
       "      <td>170</td>\n",
       "      <td>84</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.month_diff)</th>\n",
       "      <td>2951</td>\n",
       "      <td>1933</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.purchase_date_successive_diff)</th>\n",
       "      <td>2.0978e+07</td>\n",
       "      <td>1.45427e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.duration)</th>\n",
       "      <td>-1885.49</td>\n",
       "      <td>-1085.41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.amount_month_ratio)</th>\n",
       "      <td>-14.6348</td>\n",
       "      <td>-5.17162</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_hist_trans.installments)</th>\n",
       "      <td>0.123314</td>\n",
       "      <td>0.17144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_hist_trans.month_lag)</th>\n",
       "      <td>2.39769</td>\n",
       "      <td>1.40005</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_hist_trans.purchase_amount)</th>\n",
       "      <td>0.212139</td>\n",
       "      <td>0.415112</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_hist_trans.purchased_on_weekday)</th>\n",
       "      <td>0.47666</td>\n",
       "      <td>0.4842</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_hist_trans.month_diff)</th>\n",
       "      <td>0.47789</td>\n",
       "      <td>0.50074</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_hist_trans.purchase_date_successive_diff)</th>\n",
       "      <td>107376</td>\n",
       "      <td>246098</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_hist_trans.duration)</th>\n",
       "      <td>2.39005</td>\n",
       "      <td>6.13736</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_hist_trans.amount_month_ratio)</th>\n",
       "      <td>0.0191685</td>\n",
       "      <td>0.0281872</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_hist_trans.installments)</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_hist_trans.month_lag)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_hist_trans.purchase_amount)</th>\n",
       "      <td>2.25839</td>\n",
       "      <td>2.2716</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_hist_trans.purchased_on_weekday)</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_hist_trans.month_diff)</th>\n",
       "      <td>12</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_hist_trans.purchase_date_successive_diff)</th>\n",
       "      <td>1.04474e+06</td>\n",
       "      <td>1.73702e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_hist_trans.duration)</th>\n",
       "      <td>24.8423</td>\n",
       "      <td>34.074</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_hist_trans.amount_month_ratio)</th>\n",
       "      <td>0.205309</td>\n",
       "      <td>0.15144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_hist_trans.installments)</th>\n",
       "      <td>7.92077</td>\n",
       "      <td>5.56579</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_hist_trans.month_lag)</th>\n",
       "      <td>0.0659603</td>\n",
       "      <td>-0.158893</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_hist_trans.purchase_amount)</th>\n",
       "      <td>10.2431</td>\n",
       "      <td>5.23007</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_hist_trans.purchased_on_weekday)</th>\n",
       "      <td>-0.650521</td>\n",
       "      <td>-0.551787</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_hist_trans.month_diff)</th>\n",
       "      <td>0.632627</td>\n",
       "      <td>-0.137202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_hist_trans.purchase_date_successive_diff)</th>\n",
       "      <td>3.89701</td>\n",
       "      <td>4.40674</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_hist_trans.duration)</th>\n",
       "      <td>9.75999</td>\n",
       "      <td>5.31526</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_hist_trans.amount_month_ratio)</th>\n",
       "      <td>10.1906</td>\n",
       "      <td>5.09217</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_hist_trans.installments)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_hist_trans.month_lag)</th>\n",
       "      <td>-8</td>\n",
       "      <td>-5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_hist_trans.purchase_amount)</th>\n",
       "      <td>-0.739395</td>\n",
       "      <td>-0.740942</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_hist_trans.purchased_on_weekday)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_hist_trans.month_diff)</th>\n",
       "      <td>11</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_hist_trans.purchase_date_successive_diff)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_hist_trans.duration)</th>\n",
       "      <td>-8.80602</td>\n",
       "      <td>-11.0909</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_hist_trans.amount_month_ratio)</th>\n",
       "      <td>-0.0672177</td>\n",
       "      <td>-0.0529244</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.installments)</th>\n",
       "      <td>0.0153846</td>\n",
       "      <td>0.0300752</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.month_lag)</th>\n",
       "      <td>-3.91154</td>\n",
       "      <td>-2.31579</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.purchase_amount)</th>\n",
       "      <td>-0.638341</td>\n",
       "      <td>-0.562989</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.purchased_on_weekday)</th>\n",
       "      <td>0.653846</td>\n",
       "      <td>0.631579</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.month_diff)</th>\n",
       "      <td>11.35</td>\n",
       "      <td>14.5338</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.purchase_date_successive_diff)</th>\n",
       "      <td>80684.6</td>\n",
       "      <td>109344</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.duration)</th>\n",
       "      <td>-7.25187</td>\n",
       "      <td>-8.16095</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.amount_month_ratio)</th>\n",
       "      <td>-0.0562877</td>\n",
       "      <td>-0.0388844</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COUNT(temp_hist_trans)</th>\n",
       "      <td>260</td>\n",
       "      <td>133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_hist_trans.installments)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_hist_trans.month_lag)</th>\n",
       "      <td>-4</td>\n",
       "      <td>-2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_hist_trans.purchase_amount)</th>\n",
       "      <td>-0.698042</td>\n",
       "      <td>-0.701828</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_hist_trans.purchased_on_weekday)</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_hist_trans.month_diff)</th>\n",
       "      <td>11</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_hist_trans.purchase_date_successive_diff)</th>\n",
       "      <td>62157.5</td>\n",
       "      <td>29646</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_hist_trans.duration)</th>\n",
       "      <td>-7.8022</td>\n",
       "      <td>-9.95182</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_hist_trans.amount_month_ratio)</th>\n",
       "      <td>-0.0595747</td>\n",
       "      <td>-0.0469106</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_hist_trans.authorized_flag)</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_hist_trans.city_id)</th>\n",
       "      <td>7</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_hist_trans.category_1)</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_hist_trans.category_3)</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_hist_trans.merchant_category_id)</th>\n",
       "      <td>41</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_hist_trans.merchant_id)</th>\n",
       "      <td>94</td>\n",
       "      <td>74</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_hist_trans.category_2)</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_hist_trans.state_id)</th>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_hist_trans.subsector_id)</th>\n",
       "      <td>21</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_hist_trans.purchased_on_weekend)</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_hist_trans.authorized_flag)</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_hist_trans.city_id)</th>\n",
       "      <td>69</td>\n",
       "      <td>87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_hist_trans.category_1)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_hist_trans.category_3)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_hist_trans.merchant_category_id)</th>\n",
       "      <td>560</td>\n",
       "      <td>307</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_hist_trans.merchant_id)</th>\n",
       "      <td>M_ID_1a81c358a3</td>\n",
       "      <td>M_ID_c8f44ec011</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_hist_trans.category_2)</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_hist_trans.state_id)</th>\n",
       "      <td>9</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_hist_trans.subsector_id)</th>\n",
       "      <td>34</td>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_hist_trans.purchased_on_weekend)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AVG_TIME_BETWEEN(temp_hist_trans.purchase_date)</th>\n",
       "      <td>80996.1</td>\n",
       "      <td>110172</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIME_SINCE_LAST(temp_hist_trans.purchase_date)</th>\n",
       "      <td>5.07859e+07</td>\n",
       "      <td>5.19882e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_hist_trans.month_diff, purchase_date)</th>\n",
       "      <td>-2.03099e-08</td>\n",
       "      <td>-5.28998e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_hist_trans.installments, purchase_date)</th>\n",
       "      <td>-8.9272e-10</td>\n",
       "      <td>-3.35243e-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_hist_trans.amount_month_ratio, purchase_date)</th>\n",
       "      <td>3.09112e-10</td>\n",
       "      <td>-1.90272e-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_hist_trans.purchase_amount, purchase_date)</th>\n",
       "      <td>4.55973e-09</td>\n",
       "      <td>-2.61004e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_hist_trans.purchase_date_successive_diff, purchase_date)</th>\n",
       "      <td>0.00159676</td>\n",
       "      <td>0.00385725</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_hist_trans.duration, purchase_date)</th>\n",
       "      <td>6.4071e-08</td>\n",
       "      <td>-3.57342e-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_hist_trans.month_lag, purchase_date)</th>\n",
       "      <td>3.71169e-07</td>\n",
       "      <td>3.42225e-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_hist_trans.purchased_on_weekday, purchase_date)</th>\n",
       "      <td>-6.85407e-09</td>\n",
       "      <td>1.11143e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.installments)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.month_lag)</th>\n",
       "      <td>34</td>\n",
       "      <td>32</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.purchase_amount)</th>\n",
       "      <td>-13.2442</td>\n",
       "      <td>-8.18339</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.purchased_on_weekday)</th>\n",
       "      <td>17</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.month_diff)</th>\n",
       "      <td>261</td>\n",
       "      <td>342</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.purchase_date_successive_diff)</th>\n",
       "      <td>4.74231e+06</td>\n",
       "      <td>4.72866e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.duration)</th>\n",
       "      <td>-150.358</td>\n",
       "      <td>-116.269</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.amount_month_ratio)</th>\n",
       "      <td>-1.16863</td>\n",
       "      <td>-0.576425</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_new_trans.installments)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_new_trans.month_lag)</th>\n",
       "      <td>0.510754</td>\n",
       "      <td>0.481543</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_new_trans.purchase_amount)</th>\n",
       "      <td>0.135812</td>\n",
       "      <td>0.734419</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_new_trans.purchased_on_weekday)</th>\n",
       "      <td>0.448978</td>\n",
       "      <td>0.510754</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_new_trans.month_diff)</th>\n",
       "      <td>0.486985</td>\n",
       "      <td>0.442326</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_new_trans.purchase_date_successive_diff)</th>\n",
       "      <td>205878</td>\n",
       "      <td>222107</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_new_trans.duration)</th>\n",
       "      <td>1.57293</td>\n",
       "      <td>10.4416</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_new_trans.amount_month_ratio)</th>\n",
       "      <td>0.0121497</td>\n",
       "      <td>0.0517468</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_new_trans.installments)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_new_trans.month_lag)</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_new_trans.purchase_amount)</th>\n",
       "      <td>-0.296112</td>\n",
       "      <td>2.07808</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_new_trans.purchased_on_weekday)</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_new_trans.month_diff)</th>\n",
       "      <td>12</td>\n",
       "      <td>15</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_new_trans.purchase_date_successive_diff)</th>\n",
       "      <td>778975</td>\n",
       "      <td>764226</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_new_trans.duration)</th>\n",
       "      <td>-3.25724</td>\n",
       "      <td>29.0931</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_new_trans.amount_month_ratio)</th>\n",
       "      <td>-0.0256465</td>\n",
       "      <td>0.148434</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_new_trans.installments)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_new_trans.month_lag)</th>\n",
       "      <td>0.0932329</td>\n",
       "      <td>0.755148</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_new_trans.purchase_amount)</th>\n",
       "      <td>0.899623</td>\n",
       "      <td>2.51195</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_new_trans.purchased_on_weekday)</th>\n",
       "      <td>-1.16667</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_new_trans.month_diff)</th>\n",
       "      <td>0.684484</td>\n",
       "      <td>1.23315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_new_trans.purchase_date_successive_diff)</th>\n",
       "      <td>1.69853</td>\n",
       "      <td>1.28946</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_new_trans.duration)</th>\n",
       "      <td>0.857694</td>\n",
       "      <td>2.4893</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_new_trans.amount_month_ratio)</th>\n",
       "      <td>0.771362</td>\n",
       "      <td>2.5319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_new_trans.installments)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_new_trans.month_lag)</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_new_trans.purchase_amount)</th>\n",
       "      <td>-0.724368</td>\n",
       "      <td>-0.727373</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_new_trans.purchased_on_weekday)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_new_trans.month_diff)</th>\n",
       "      <td>11</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_new_trans.purchase_date_successive_diff)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_new_trans.duration)</th>\n",
       "      <td>-8.63868</td>\n",
       "      <td>-10.5725</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_new_trans.amount_month_ratio)</th>\n",
       "      <td>-0.0658516</td>\n",
       "      <td>-0.0519552</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.installments)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.month_lag)</th>\n",
       "      <td>1.47826</td>\n",
       "      <td>1.33333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.purchase_amount)</th>\n",
       "      <td>-0.575835</td>\n",
       "      <td>-0.340975</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.purchased_on_weekday)</th>\n",
       "      <td>0.73913</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.month_diff)</th>\n",
       "      <td>11.3478</td>\n",
       "      <td>14.25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.purchase_date_successive_diff)</th>\n",
       "      <td>206187</td>\n",
       "      <td>197027</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.duration)</th>\n",
       "      <td>-6.53729</td>\n",
       "      <td>-4.84455</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.amount_month_ratio)</th>\n",
       "      <td>-0.0508099</td>\n",
       "      <td>-0.0240177</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>COUNT(temp_new_trans)</th>\n",
       "      <td>23</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_new_trans.installments)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_new_trans.month_lag)</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_new_trans.purchase_amount)</th>\n",
       "      <td>-0.58118</td>\n",
       "      <td>-0.609438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_new_trans.purchased_on_weekday)</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_new_trans.month_diff)</th>\n",
       "      <td>11</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_new_trans.purchase_date_successive_diff)</th>\n",
       "      <td>164976</td>\n",
       "      <td>119832</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_new_trans.duration)</th>\n",
       "      <td>-6.83875</td>\n",
       "      <td>-8.53213</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_new_trans.amount_month_ratio)</th>\n",
       "      <td>-0.0488853</td>\n",
       "      <td>-0.0435313</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_new_trans.authorized_flag)</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_new_trans.city_id)</th>\n",
       "      <td>3</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_new_trans.category_1)</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_new_trans.category_3)</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_new_trans.merchant_category_id)</th>\n",
       "      <td>14</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_new_trans.merchant_id)</th>\n",
       "      <td>23</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_new_trans.category_2)</th>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_new_trans.state_id)</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_new_trans.subsector_id)</th>\n",
       "      <td>10</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_new_trans.purchased_on_weekend)</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_new_trans.authorized_flag)</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_new_trans.city_id)</th>\n",
       "      <td>69</td>\n",
       "      <td>87</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_new_trans.category_1)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_new_trans.category_3)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_new_trans.merchant_category_id)</th>\n",
       "      <td>278</td>\n",
       "      <td>278</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_new_trans.merchant_id)</th>\n",
       "      <td>M_ID_0703c10cc0</td>\n",
       "      <td>M_ID_00a6ca8a8a</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_new_trans.category_2)</th>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_new_trans.state_id)</th>\n",
       "      <td>9</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_new_trans.subsector_id)</th>\n",
       "      <td>37</td>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_new_trans.purchased_on_weekend)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>AVG_TIME_BETWEEN(temp_new_trans.purchase_date)</th>\n",
       "      <td>215560</td>\n",
       "      <td>205594</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TIME_SINCE_LAST(temp_new_trans.purchase_date)</th>\n",
       "      <td>2.91004e+07</td>\n",
       "      <td>3.71267e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_new_trans.installments, purchase_date)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_new_trans.amount_month_ratio, purchase_date)</th>\n",
       "      <td>-1.46783e-09</td>\n",
       "      <td>-1.94178e-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_new_trans.purchase_amount, purchase_date)</th>\n",
       "      <td>-1.10757e-08</td>\n",
       "      <td>-3.03324e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_new_trans.purchase_date_successive_diff, purchase_date)</th>\n",
       "      <td>0.03212</td>\n",
       "      <td>0.0588533</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_new_trans.duration, purchase_date)</th>\n",
       "      <td>-6.09874e-08</td>\n",
       "      <td>-4.71864e-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_new_trans.purchased_on_weekday, purchase_date)</th>\n",
       "      <td>-5.64721e-08</td>\n",
       "      <td>1.27364e-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_new_trans.month_diff, purchase_date)</th>\n",
       "      <td>-8.02519e-08</td>\n",
       "      <td>-7.76171e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_new_trans.month_lag, purchase_date)</th>\n",
       "      <td>2.96858e-07</td>\n",
       "      <td>3.27168e-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DAY(first_active_month)</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MONTH(first_active_month)</th>\n",
       "      <td>6</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>WEEKDAY(first_active_month)</th>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DAYS_SINCE(first_active_month)</th>\n",
       "      <td>614</td>\n",
       "      <td>614</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(target)</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>HOUR(first_active_month)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.installments WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.installments WHERE authorized_flag = 1)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.month_lag WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>-326</td>\n",
       "      <td>-122</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.month_lag WHERE authorized_flag = 1)</th>\n",
       "      <td>-959</td>\n",
       "      <td>-263</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.purchase_amount WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>-55.8855</td>\n",
       "      <td>-27.0708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.purchase_amount WHERE authorized_flag = 1)</th>\n",
       "      <td>-157.397</td>\n",
       "      <td>-69.2762</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.purchased_on_weekday WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.purchased_on_weekday WHERE authorized_flag = 1)</th>\n",
       "      <td>161</td>\n",
       "      <td>70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.month_diff WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>1017</td>\n",
       "      <td>713</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.month_diff WHERE authorized_flag = 1)</th>\n",
       "      <td>2802</td>\n",
       "      <td>1674</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.purchase_date_successive_diff WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>7.96363e+06</td>\n",
       "      <td>3.9765e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.purchase_date_successive_diff WHERE authorized_flag = 1)</th>\n",
       "      <td>1.94592e+07</td>\n",
       "      <td>1.27357e+07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.duration WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>-630.755</td>\n",
       "      <td>-392.683</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.duration WHERE authorized_flag = 1)</th>\n",
       "      <td>-1787.31</td>\n",
       "      <td>-1007.95</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.amount_month_ratio WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>-4.95917</td>\n",
       "      <td>-1.86843</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.amount_month_ratio WHERE authorized_flag = 1)</th>\n",
       "      <td>-13.885</td>\n",
       "      <td>-4.76694</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.DAYS_SINCE(purchase_date))</th>\n",
       "      <td>123197</td>\n",
       "      <td>69124</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.DAYS_SINCE(purchase_date) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>41802</td>\n",
       "      <td>25744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.DAYS_SINCE(purchase_date) WHERE authorized_flag = 1)</th>\n",
       "      <td>116792</td>\n",
       "      <td>59684</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(installments))</th>\n",
       "      <td>129.684</td>\n",
       "      <td>67.3155</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(installments) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>44.6985</td>\n",
       "      <td>24.5636</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(installments) WHERE authorized_flag = 1)</th>\n",
       "      <td>121.3</td>\n",
       "      <td>56.4758</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(month_lag))</th>\n",
       "      <td>113.419</td>\n",
       "      <td>83.5814</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(month_lag) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>42.1234</td>\n",
       "      <td>29.458</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(month_lag) WHERE authorized_flag = 1)</th>\n",
       "      <td>108.859</td>\n",
       "      <td>72.7519</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(purchase_amount))</th>\n",
       "      <td>130.022</td>\n",
       "      <td>66.9784</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(purchase_amount) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>51.9046</td>\n",
       "      <td>29.1539</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(purchase_amount) WHERE authorized_flag = 1)</th>\n",
       "      <td>124.229</td>\n",
       "      <td>59.6221</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(purchased_on_weekday))</th>\n",
       "      <td>131.31</td>\n",
       "      <td>65.6896</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(purchased_on_weekday) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>16.0305</td>\n",
       "      <td>8.72774</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(purchased_on_weekday) WHERE authorized_flag = 1)</th>\n",
       "      <td>124.495</td>\n",
       "      <td>55.4835</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(month_diff))</th>\n",
       "      <td>86.3359</td>\n",
       "      <td>110.664</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(month_diff) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>28.3969</td>\n",
       "      <td>40.9135</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(month_diff) WHERE authorized_flag = 1)</th>\n",
       "      <td>81.5394</td>\n",
       "      <td>96.1285</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(purchase_date_successive_diff))</th>\n",
       "      <td>135.628</td>\n",
       "      <td>61.3715</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(purchase_date_successive_diff) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>46.6476</td>\n",
       "      <td>20.2608</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(purchase_date_successive_diff) WHERE authorized_flag = 1)</th>\n",
       "      <td>129.425</td>\n",
       "      <td>52.5216</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(duration))</th>\n",
       "      <td>158.537</td>\n",
       "      <td>38.4631</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(duration) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>61.1387</td>\n",
       "      <td>18.4924</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(duration) WHERE authorized_flag = 1)</th>\n",
       "      <td>151.294</td>\n",
       "      <td>33.0534</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(amount_month_ratio))</th>\n",
       "      <td>96.2774</td>\n",
       "      <td>100.723</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(amount_month_ratio) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>38.3702</td>\n",
       "      <td>38.8664</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_hist_trans.PERCENTILE(amount_month_ratio) WHERE authorized_flag = 1)</th>\n",
       "      <td>91.6578</td>\n",
       "      <td>87.7735</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_hist_trans.DAYS_SINCE(purchase_date))</th>\n",
       "      <td>74.2452</td>\n",
       "      <td>46.4585</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_hist_trans.PERCENTILE(installments))</th>\n",
       "      <td>0.0616571</td>\n",
       "      <td>0.08572</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_hist_trans.PERCENTILE(month_lag))</th>\n",
       "      <td>0.2978</td>\n",
       "      <td>0.208161</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_hist_trans.PERCENTILE(purchase_amount))</th>\n",
       "      <td>0.273598</td>\n",
       "      <td>0.318093</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_hist_trans.PERCENTILE(purchased_on_weekday))</th>\n",
       "      <td>0.23833</td>\n",
       "      <td>0.2421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_hist_trans.PERCENTILE(month_diff))</th>\n",
       "      <td>0.15808</td>\n",
       "      <td>0.0847308</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_hist_trans.PERCENTILE(purchase_date_successive_diff))</th>\n",
       "      <td>0.280444</td>\n",
       "      <td>0.302273</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_hist_trans.PERCENTILE(duration))</th>\n",
       "      <td>0.20574</td>\n",
       "      <td>0.310715</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_hist_trans.PERCENTILE(amount_month_ratio))</th>\n",
       "      <td>0.258088</td>\n",
       "      <td>0.132508</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_hist_trans.DAYS_SINCE(purchase_date))</th>\n",
       "      <td>587</td>\n",
       "      <td>601</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_hist_trans.PERCENTILE(installments))</th>\n",
       "      <td>0.991094</td>\n",
       "      <td>0.991094</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_hist_trans.PERCENTILE(month_lag))</th>\n",
       "      <td>0.959288</td>\n",
       "      <td>0.959288</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_hist_trans.PERCENTILE(purchase_amount))</th>\n",
       "      <td>0.994911</td>\n",
       "      <td>0.998728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_hist_trans.PERCENTILE(purchased_on_weekday))</th>\n",
       "      <td>0.678117</td>\n",
       "      <td>0.678117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_hist_trans.PERCENTILE(month_diff))</th>\n",
       "      <td>0.547074</td>\n",
       "      <td>0.910941</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_hist_trans.PERCENTILE(purchase_date_successive_diff))</th>\n",
       "      <td>0.992366</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_hist_trans.PERCENTILE(duration))</th>\n",
       "      <td>0.994911</td>\n",
       "      <td>0.998728</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_hist_trans.PERCENTILE(amount_month_ratio))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.996183</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_hist_trans.DAYS_SINCE(purchase_date))</th>\n",
       "      <td>-0.0534497</td>\n",
       "      <td>0.0429153</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_hist_trans.PERCENTILE(installments))</th>\n",
       "      <td>7.92077</td>\n",
       "      <td>5.56579</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_hist_trans.PERCENTILE(month_lag))</th>\n",
       "      <td>0.290522</td>\n",
       "      <td>-0.0287405</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_hist_trans.PERCENTILE(purchase_amount))</th>\n",
       "      <td>0.0065437</td>\n",
       "      <td>-0.0135877</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_hist_trans.PERCENTILE(purchased_on_weekday))</th>\n",
       "      <td>-0.650521</td>\n",
       "      <td>-0.551787</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_hist_trans.PERCENTILE(month_diff))</th>\n",
       "      <td>0.632627</td>\n",
       "      <td>-0.137202</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_hist_trans.PERCENTILE(purchase_date_successive_diff))</th>\n",
       "      <td>-0.158507</td>\n",
       "      <td>0.306568</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_hist_trans.PERCENTILE(duration))</th>\n",
       "      <td>0.00399592</td>\n",
       "      <td>1.40217</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_hist_trans.PERCENTILE(amount_month_ratio))</th>\n",
       "      <td>0.745192</td>\n",
       "      <td>0.189281</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_hist_trans.DAYS_SINCE(purchase_date))</th>\n",
       "      <td>344</td>\n",
       "      <td>433</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_hist_trans.PERCENTILE(installments))</th>\n",
       "      <td>0.491094</td>\n",
       "      <td>0.491094</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_hist_trans.PERCENTILE(month_lag))</th>\n",
       "      <td>0.00508906</td>\n",
       "      <td>0.276081</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_hist_trans.PERCENTILE(purchase_amount))</th>\n",
       "      <td>0.0114504</td>\n",
       "      <td>0.00381679</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_hist_trans.PERCENTILE(purchased_on_weekday))</th>\n",
       "      <td>0.178117</td>\n",
       "      <td>0.178117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_hist_trans.PERCENTILE(month_diff))</th>\n",
       "      <td>0.216285</td>\n",
       "      <td>0.74173</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_hist_trans.PERCENTILE(purchase_date_successive_diff))</th>\n",
       "      <td>0.00508906</td>\n",
       "      <td>0.00508906</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_hist_trans.PERCENTILE(duration))</th>\n",
       "      <td>0.246819</td>\n",
       "      <td>0.00254453</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_hist_trans.PERCENTILE(amount_month_ratio))</th>\n",
       "      <td>0.00254453</td>\n",
       "      <td>0.533079</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.installments WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0.0111111</td>\n",
       "      <td>0.0204082</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.installments WHERE authorized_flag = 1)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.month_lag WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>-3.62222</td>\n",
       "      <td>-2.4898</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.month_lag WHERE authorized_flag = 1)</th>\n",
       "      <td>-3.88259</td>\n",
       "      <td>-2.28696</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.purchase_amount WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>-0.62095</td>\n",
       "      <td>-0.552466</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.purchase_amount WHERE authorized_flag = 1)</th>\n",
       "      <td>-0.637235</td>\n",
       "      <td>-0.602401</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.purchased_on_weekday WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.purchased_on_weekday WHERE authorized_flag = 1)</th>\n",
       "      <td>0.651822</td>\n",
       "      <td>0.608696</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.month_diff WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>11.3</td>\n",
       "      <td>14.551</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.month_diff WHERE authorized_flag = 1)</th>\n",
       "      <td>11.3441</td>\n",
       "      <td>14.5565</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.purchase_date_successive_diff WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>88484.8</td>\n",
       "      <td>81153.1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.purchase_date_successive_diff WHERE authorized_flag = 1)</th>\n",
       "      <td>78782.3</td>\n",
       "      <td>110745</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.duration WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>-7.00839</td>\n",
       "      <td>-8.01394</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.duration WHERE authorized_flag = 1)</th>\n",
       "      <td>-7.23609</td>\n",
       "      <td>-8.7648</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.amount_month_ratio WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>-0.0551019</td>\n",
       "      <td>-0.0381313</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.amount_month_ratio WHERE authorized_flag = 1)</th>\n",
       "      <td>-0.0562145</td>\n",
       "      <td>-0.0414517</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.DAYS_SINCE(purchase_date))</th>\n",
       "      <td>473.835</td>\n",
       "      <td>519.729</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.DAYS_SINCE(purchase_date) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>464.467</td>\n",
       "      <td>525.388</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.DAYS_SINCE(purchase_date) WHERE authorized_flag = 1)</th>\n",
       "      <td>472.842</td>\n",
       "      <td>518.991</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(installments))</th>\n",
       "      <td>0.498786</td>\n",
       "      <td>0.506132</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(installments) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0.49665</td>\n",
       "      <td>0.501298</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(installments) WHERE authorized_flag = 1)</th>\n",
       "      <td>0.491094</td>\n",
       "      <td>0.491094</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(month_lag))</th>\n",
       "      <td>0.436225</td>\n",
       "      <td>0.628432</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(month_lag) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0.468038</td>\n",
       "      <td>0.601184</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(month_lag) WHERE authorized_flag = 1)</th>\n",
       "      <td>0.440724</td>\n",
       "      <td>0.632625</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(purchase_amount))</th>\n",
       "      <td>0.500083</td>\n",
       "      <td>0.503597</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(purchase_amount) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0.576718</td>\n",
       "      <td>0.594978</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(purchase_amount) WHERE authorized_flag = 1)</th>\n",
       "      <td>0.502951</td>\n",
       "      <td>0.518453</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(purchased_on_weekday))</th>\n",
       "      <td>0.50504</td>\n",
       "      <td>0.493907</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(purchased_on_weekday) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0.178117</td>\n",
       "      <td>0.178117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(purchased_on_weekday) WHERE authorized_flag = 1)</th>\n",
       "      <td>0.504028</td>\n",
       "      <td>0.482465</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(month_diff))</th>\n",
       "      <td>0.332061</td>\n",
       "      <td>0.832061</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(month_diff) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0.315522</td>\n",
       "      <td>0.834969</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(month_diff) WHERE authorized_flag = 1)</th>\n",
       "      <td>0.330119</td>\n",
       "      <td>0.8359</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(purchase_date_successive_diff))</th>\n",
       "      <td>0.521648</td>\n",
       "      <td>0.46144</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(purchase_date_successive_diff) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0.518306</td>\n",
       "      <td>0.413486</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(purchase_date_successive_diff) WHERE authorized_flag = 1)</th>\n",
       "      <td>0.523988</td>\n",
       "      <td>0.45671</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(duration))</th>\n",
       "      <td>0.609757</td>\n",
       "      <td>0.289196</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(duration) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0.679319</td>\n",
       "      <td>0.377395</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(duration) WHERE authorized_flag = 1)</th>\n",
       "      <td>0.612526</td>\n",
       "      <td>0.287421</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(amount_month_ratio))</th>\n",
       "      <td>0.370298</td>\n",
       "      <td>0.757313</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(amount_month_ratio) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0.426336</td>\n",
       "      <td>0.793192</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_hist_trans.PERCENTILE(amount_month_ratio) WHERE authorized_flag = 1)</th>\n",
       "      <td>0.371084</td>\n",
       "      <td>0.763248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_hist_trans.DAYS_SINCE(purchase_date))</th>\n",
       "      <td>475.5</td>\n",
       "      <td>514</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_hist_trans.PERCENTILE(installments))</th>\n",
       "      <td>0.491094</td>\n",
       "      <td>0.491094</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_hist_trans.PERCENTILE(month_lag))</th>\n",
       "      <td>0.382952</td>\n",
       "      <td>0.655216</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_hist_trans.PERCENTILE(purchase_amount))</th>\n",
       "      <td>0.510178</td>\n",
       "      <td>0.466921</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_hist_trans.PERCENTILE(purchased_on_weekday))</th>\n",
       "      <td>0.678117</td>\n",
       "      <td>0.678117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_hist_trans.PERCENTILE(month_diff))</th>\n",
       "      <td>0.216285</td>\n",
       "      <td>0.910941</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_hist_trans.PERCENTILE(purchase_date_successive_diff))</th>\n",
       "      <td>0.553435</td>\n",
       "      <td>0.440204</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_hist_trans.PERCENTILE(duration))</th>\n",
       "      <td>0.60687</td>\n",
       "      <td>0.170483</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_hist_trans.PERCENTILE(amount_month_ratio))</th>\n",
       "      <td>0.332061</td>\n",
       "      <td>0.745547</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_hist_trans.DAY(purchase_date))</th>\n",
       "      <td>31</td>\n",
       "      <td>27</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_hist_trans.MONTH(purchase_date))</th>\n",
       "      <td>9</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_hist_trans.WEEKDAY(purchase_date))</th>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_hist_trans.HOUR(purchase_date))</th>\n",
       "      <td>23</td>\n",
       "      <td>22</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_hist_trans.DAY(purchase_date))</th>\n",
       "      <td>11</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_hist_trans.MONTH(purchase_date))</th>\n",
       "      <td>12</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_hist_trans.WEEKDAY(purchase_date))</th>\n",
       "      <td>5</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_hist_trans.HOUR(purchase_date))</th>\n",
       "      <td>14</td>\n",
       "      <td>20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_hist_trans.PERCENTILE(month_lag), purchase_date)</th>\n",
       "      <td>4.58304e-08</td>\n",
       "      <td>5.09193e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_hist_trans.PERCENTILE(duration), purchase_date)</th>\n",
       "      <td>8.4611e-09</td>\n",
       "      <td>-1.8733e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_hist_trans.PERCENTILE(amount_month_ratio), purchase_date)</th>\n",
       "      <td>4.19797e-09</td>\n",
       "      <td>-1.29852e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_hist_trans.PERCENTILE(purchase_amount), purchase_date)</th>\n",
       "      <td>6.24958e-09</td>\n",
       "      <td>-2.40576e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_hist_trans.DAYS_SINCE(purchase_date), purchase_date)</th>\n",
       "      <td>-1.15739e-05</td>\n",
       "      <td>-1.15688e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_hist_trans.PERCENTILE(installments), purchase_date)</th>\n",
       "      <td>-4.4636e-10</td>\n",
       "      <td>-1.67622e-10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_hist_trans.PERCENTILE(purchased_on_weekday), purchase_date)</th>\n",
       "      <td>-3.42703e-09</td>\n",
       "      <td>5.55714e-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_hist_trans.PERCENTILE(purchase_date_successive_diff), purchase_date)</th>\n",
       "      <td>2.49699e-09</td>\n",
       "      <td>9.76888e-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_hist_trans.PERCENTILE(month_diff), purchase_date)</th>\n",
       "      <td>-6.71828e-09</td>\n",
       "      <td>-8.95123e-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.installments WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.month_lag WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>9</td>\n",
       "      <td>14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.purchase_amount WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>-2.81409</td>\n",
       "      <td>-4.59485</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.purchased_on_weekday WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.month_diff WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>67</td>\n",
       "      <td>171</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.purchase_date_successive_diff WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>1.48362e+06</td>\n",
       "      <td>1.85167e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.duration WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>-31.2627</td>\n",
       "      <td>-65.8502</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.amount_month_ratio WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>-0.253495</td>\n",
       "      <td>-0.320955</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.DAYS_SINCE(purchase_date))</th>\n",
       "      <td>7125</td>\n",
       "      <td>9703</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.DAYS_SINCE(purchase_date) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>1829</td>\n",
       "      <td>4905</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.PERCENTILE(installments))</th>\n",
       "      <td>11.7447</td>\n",
       "      <td>12.2553</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.PERCENTILE(installments) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>3.06383</td>\n",
       "      <td>6.12766</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.PERCENTILE(month_lag))</th>\n",
       "      <td>12.5957</td>\n",
       "      <td>11.4043</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.PERCENTILE(month_lag) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>3.35106</td>\n",
       "      <td>4.70213</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.PERCENTILE(purchase_amount))</th>\n",
       "      <td>11.6277</td>\n",
       "      <td>12.3723</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.PERCENTILE(purchase_amount) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>4.28723</td>\n",
       "      <td>5.51064</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.PERCENTILE(purchased_on_weekday))</th>\n",
       "      <td>13.1489</td>\n",
       "      <td>10.8511</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.PERCENTILE(purchased_on_weekday) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>1.21277</td>\n",
       "      <td>2.42553</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.PERCENTILE(month_diff))</th>\n",
       "      <td>5.87234</td>\n",
       "      <td>18.1277</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.PERCENTILE(month_diff) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>1.26596</td>\n",
       "      <td>9.06383</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.PERCENTILE(purchase_date_successive_diff))</th>\n",
       "      <td>12.3511</td>\n",
       "      <td>11.6489</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.PERCENTILE(purchase_date_successive_diff) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>3.68085</td>\n",
       "      <td>4.90426</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.PERCENTILE(duration))</th>\n",
       "      <td>14.4043</td>\n",
       "      <td>9.59574</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.PERCENTILE(duration) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>4.71277</td>\n",
       "      <td>4.34043</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.PERCENTILE(amount_month_ratio))</th>\n",
       "      <td>8.80851</td>\n",
       "      <td>15.1915</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SUM(temp_new_trans.PERCENTILE(amount_month_ratio) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>3.62766</td>\n",
       "      <td>6.97872</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_new_trans.DAYS_SINCE(purchase_date))</th>\n",
       "      <td>16.6022</td>\n",
       "      <td>14.5018</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_new_trans.PERCENTILE(installments))</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_new_trans.PERCENTILE(month_lag))</th>\n",
       "      <td>0.255377</td>\n",
       "      <td>0.240772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_new_trans.PERCENTILE(purchase_amount))</th>\n",
       "      <td>0.271089</td>\n",
       "      <td>0.31591</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_new_trans.PERCENTILE(purchased_on_weekday))</th>\n",
       "      <td>0.224489</td>\n",
       "      <td>0.255377</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_new_trans.PERCENTILE(month_diff))</th>\n",
       "      <td>0.119156</td>\n",
       "      <td>0.112934</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_new_trans.PERCENTILE(purchase_date_successive_diff))</th>\n",
       "      <td>0.26951</td>\n",
       "      <td>0.315221</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_new_trans.PERCENTILE(duration))</th>\n",
       "      <td>0.199755</td>\n",
       "      <td>0.325332</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>STD(temp_new_trans.PERCENTILE(amount_month_ratio))</th>\n",
       "      <td>0.286876</td>\n",
       "      <td>0.244047</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_new_trans.DAYS_SINCE(purchase_date))</th>\n",
       "      <td>336</td>\n",
       "      <td>429</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_new_trans.PERCENTILE(installments))</th>\n",
       "      <td>0.510638</td>\n",
       "      <td>0.510638</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_new_trans.PERCENTILE(month_lag))</th>\n",
       "      <td>0.808511</td>\n",
       "      <td>0.808511</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_new_trans.PERCENTILE(purchase_amount))</th>\n",
       "      <td>0.904255</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_new_trans.PERCENTILE(purchased_on_weekday))</th>\n",
       "      <td>0.702128</td>\n",
       "      <td>0.702128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_new_trans.PERCENTILE(month_diff))</th>\n",
       "      <td>0.414894</td>\n",
       "      <td>0.946809</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_new_trans.PERCENTILE(purchase_date_successive_diff))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.978723</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_new_trans.PERCENTILE(duration))</th>\n",
       "      <td>0.925532</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MAX(temp_new_trans.PERCENTILE(amount_month_ratio))</th>\n",
       "      <td>0.914894</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_new_trans.DAYS_SINCE(purchase_date))</th>\n",
       "      <td>0.0263921</td>\n",
       "      <td>-0.217544</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_new_trans.PERCENTILE(installments))</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_new_trans.PERCENTILE(month_lag))</th>\n",
       "      <td>0.0932329</td>\n",
       "      <td>0.755148</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_new_trans.PERCENTILE(purchase_amount))</th>\n",
       "      <td>-0.113279</td>\n",
       "      <td>0.0557786</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_new_trans.PERCENTILE(purchased_on_weekday))</th>\n",
       "      <td>-1.16667</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_new_trans.PERCENTILE(month_diff))</th>\n",
       "      <td>0.684484</td>\n",
       "      <td>1.23315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_new_trans.PERCENTILE(purchase_date_successive_diff))</th>\n",
       "      <td>-0.127311</td>\n",
       "      <td>0.147175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_new_trans.PERCENTILE(duration))</th>\n",
       "      <td>-0.128077</td>\n",
       "      <td>0.719544</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>SKEW(temp_new_trans.PERCENTILE(amount_month_ratio))</th>\n",
       "      <td>0.50927</td>\n",
       "      <td>-0.153612</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_new_trans.DAYS_SINCE(purchase_date))</th>\n",
       "      <td>281</td>\n",
       "      <td>374</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_new_trans.PERCENTILE(installments))</th>\n",
       "      <td>0.510638</td>\n",
       "      <td>0.510638</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_new_trans.PERCENTILE(month_lag))</th>\n",
       "      <td>0.308511</td>\n",
       "      <td>0.308511</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_new_trans.PERCENTILE(purchase_amount))</th>\n",
       "      <td>0.0638298</td>\n",
       "      <td>0.0212766</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_new_trans.PERCENTILE(purchased_on_weekday))</th>\n",
       "      <td>0.202128</td>\n",
       "      <td>0.202128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_new_trans.PERCENTILE(month_diff))</th>\n",
       "      <td>0.170213</td>\n",
       "      <td>0.691489</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_new_trans.PERCENTILE(purchase_date_successive_diff))</th>\n",
       "      <td>0.0319149</td>\n",
       "      <td>0.0319149</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_new_trans.PERCENTILE(duration))</th>\n",
       "      <td>0.276596</td>\n",
       "      <td>0.0212766</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MIN(temp_new_trans.PERCENTILE(amount_month_ratio))</th>\n",
       "      <td>0.0212766</td>\n",
       "      <td>0.255319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.installments WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.month_lag WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>1.5</td>\n",
       "      <td>1.16667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.purchase_amount WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>-0.469015</td>\n",
       "      <td>-0.382904</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.purchased_on_weekday WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.month_diff WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>11.1667</td>\n",
       "      <td>14.25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.purchase_date_successive_diff WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>247271</td>\n",
       "      <td>154306</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.duration WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>-5.21046</td>\n",
       "      <td>-5.48752</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.amount_month_ratio WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>-0.0422491</td>\n",
       "      <td>-0.0267462</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.DAYS_SINCE(purchase_date))</th>\n",
       "      <td>309.783</td>\n",
       "      <td>404.292</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.DAYS_SINCE(purchase_date) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>304.833</td>\n",
       "      <td>408.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.PERCENTILE(installments))</th>\n",
       "      <td>0.510638</td>\n",
       "      <td>0.510638</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.PERCENTILE(installments) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0.510638</td>\n",
       "      <td>0.510638</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.PERCENTILE(month_lag))</th>\n",
       "      <td>0.547641</td>\n",
       "      <td>0.475177</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.PERCENTILE(month_lag) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0.558511</td>\n",
       "      <td>0.391844</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.PERCENTILE(purchase_amount))</th>\n",
       "      <td>0.50555</td>\n",
       "      <td>0.515514</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.PERCENTILE(purchase_amount) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0.714539</td>\n",
       "      <td>0.45922</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.PERCENTILE(purchased_on_weekday))</th>\n",
       "      <td>0.571693</td>\n",
       "      <td>0.452128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.PERCENTILE(purchased_on_weekday) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0.202128</td>\n",
       "      <td>0.202128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.PERCENTILE(month_diff))</th>\n",
       "      <td>0.255319</td>\n",
       "      <td>0.755319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.PERCENTILE(month_diff) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0.210993</td>\n",
       "      <td>0.755319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.PERCENTILE(purchase_date_successive_diff))</th>\n",
       "      <td>0.537003</td>\n",
       "      <td>0.485372</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.PERCENTILE(purchase_date_successive_diff) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0.613475</td>\n",
       "      <td>0.408688</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.PERCENTILE(duration))</th>\n",
       "      <td>0.626272</td>\n",
       "      <td>0.399823</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.PERCENTILE(duration) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0.785461</td>\n",
       "      <td>0.361702</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.PERCENTILE(amount_month_ratio))</th>\n",
       "      <td>0.382979</td>\n",
       "      <td>0.632979</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEAN(temp_new_trans.PERCENTILE(amount_month_ratio) WHERE purchased_on_weekend = 1)</th>\n",
       "      <td>0.60461</td>\n",
       "      <td>0.58156</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_new_trans.DAYS_SINCE(purchase_date))</th>\n",
       "      <td>310</td>\n",
       "      <td>407</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_new_trans.PERCENTILE(installments))</th>\n",
       "      <td>0.510638</td>\n",
       "      <td>0.510638</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_new_trans.PERCENTILE(month_lag))</th>\n",
       "      <td>0.308511</td>\n",
       "      <td>0.308511</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_new_trans.PERCENTILE(purchase_amount))</th>\n",
       "      <td>0.553191</td>\n",
       "      <td>0.484043</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_new_trans.PERCENTILE(purchased_on_weekday))</th>\n",
       "      <td>0.702128</td>\n",
       "      <td>0.452128</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_new_trans.PERCENTILE(month_diff))</th>\n",
       "      <td>0.170213</td>\n",
       "      <td>0.691489</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_new_trans.PERCENTILE(purchase_date_successive_diff))</th>\n",
       "      <td>0.553191</td>\n",
       "      <td>0.468085</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_new_trans.PERCENTILE(duration))</th>\n",
       "      <td>0.638298</td>\n",
       "      <td>0.287234</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MEDIAN(temp_new_trans.PERCENTILE(amount_month_ratio))</th>\n",
       "      <td>0.404255</td>\n",
       "      <td>0.680851</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_new_trans.DAY(purchase_date))</th>\n",
       "      <td>17</td>\n",
       "      <td>17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_new_trans.MONTH(purchase_date))</th>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_new_trans.WEEKDAY(purchase_date))</th>\n",
       "      <td>7</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>NUM_UNIQUE(temp_new_trans.HOUR(purchase_date))</th>\n",
       "      <td>8</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_new_trans.DAY(purchase_date))</th>\n",
       "      <td>6</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_new_trans.MONTH(purchase_date))</th>\n",
       "      <td>3</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_new_trans.WEEKDAY(purchase_date))</th>\n",
       "      <td>4</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>MODE(temp_new_trans.HOUR(purchase_date))</th>\n",
       "      <td>13</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_new_trans.PERCENTILE(purchased_on_weekday), purchase_date)</th>\n",
       "      <td>-2.8236e-08</td>\n",
       "      <td>6.36819e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_new_trans.PERCENTILE(duration), purchase_date)</th>\n",
       "      <td>1.50261e-09</td>\n",
       "      <td>5.40778e-09</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_new_trans.PERCENTILE(month_diff), purchase_date)</th>\n",
       "      <td>-1.96361e-08</td>\n",
       "      <td>-1.98171e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_new_trans.PERCENTILE(amount_month_ratio), purchase_date)</th>\n",
       "      <td>-2.25804e-08</td>\n",
       "      <td>2.4331e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_new_trans.PERCENTILE(installments), purchase_date)</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_new_trans.DAYS_SINCE(purchase_date), purchase_date)</th>\n",
       "      <td>-1.15877e-05</td>\n",
       "      <td>-1.15842e-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_new_trans.PERCENTILE(purchase_amount), purchase_date)</th>\n",
       "      <td>-1.18799e-08</td>\n",
       "      <td>1.73688e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_new_trans.PERCENTILE(month_lag), purchase_date)</th>\n",
       "      <td>1.48429e-07</td>\n",
       "      <td>1.63584e-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>TREND(temp_new_trans.PERCENTILE(purchase_date_successive_diff), purchase_date)</th>\n",
       "      <td>3.6465e-08</td>\n",
       "      <td>8.74523e-08</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SUM(temp_hist_trans.installments))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SUM(temp_hist_trans.month_lag))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SUM(temp_hist_trans.purchase_amount))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SUM(temp_hist_trans.purchased_on_weekday))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SUM(temp_hist_trans.month_diff))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SUM(temp_hist_trans.purchase_date_successive_diff))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SUM(temp_hist_trans.duration))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SUM(temp_hist_trans.amount_month_ratio))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(STD(temp_hist_trans.installments))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(STD(temp_hist_trans.month_lag))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(STD(temp_hist_trans.purchase_amount))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(STD(temp_hist_trans.purchased_on_weekday))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(STD(temp_hist_trans.month_diff))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(STD(temp_hist_trans.purchase_date_successive_diff))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(STD(temp_hist_trans.duration))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(STD(temp_hist_trans.amount_month_ratio))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MAX(temp_hist_trans.installments))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MAX(temp_hist_trans.month_lag))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MAX(temp_hist_trans.purchase_amount))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MAX(temp_hist_trans.purchased_on_weekday))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MAX(temp_hist_trans.month_diff))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MAX(temp_hist_trans.purchase_date_successive_diff))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MAX(temp_hist_trans.duration))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MAX(temp_hist_trans.amount_month_ratio))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SKEW(temp_hist_trans.installments))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SKEW(temp_hist_trans.month_lag))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SKEW(temp_hist_trans.purchase_amount))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SKEW(temp_hist_trans.purchased_on_weekday))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SKEW(temp_hist_trans.month_diff))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SKEW(temp_hist_trans.purchase_date_successive_diff))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SKEW(temp_hist_trans.duration))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SKEW(temp_hist_trans.amount_month_ratio))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MIN(temp_hist_trans.installments))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MIN(temp_hist_trans.month_lag))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MIN(temp_hist_trans.purchase_amount))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MIN(temp_hist_trans.purchased_on_weekday))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MIN(temp_hist_trans.month_diff))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MIN(temp_hist_trans.purchase_date_successive_diff))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MIN(temp_hist_trans.duration))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MIN(temp_hist_trans.amount_month_ratio))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEAN(temp_hist_trans.installments))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEAN(temp_hist_trans.month_lag))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEAN(temp_hist_trans.purchase_amount))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEAN(temp_hist_trans.purchased_on_weekday))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEAN(temp_hist_trans.month_diff))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEAN(temp_hist_trans.purchase_date_successive_diff))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEAN(temp_hist_trans.duration))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEAN(temp_hist_trans.amount_month_ratio))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(COUNT(temp_hist_trans))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEDIAN(temp_hist_trans.installments))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEDIAN(temp_hist_trans.month_lag))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEDIAN(temp_hist_trans.purchase_amount))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEDIAN(temp_hist_trans.purchased_on_weekday))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEDIAN(temp_hist_trans.month_diff))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEDIAN(temp_hist_trans.purchase_date_successive_diff))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEDIAN(temp_hist_trans.duration))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEDIAN(temp_hist_trans.amount_month_ratio))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_hist_trans.authorized_flag))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_hist_trans.city_id))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_hist_trans.category_1))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_hist_trans.category_3))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_hist_trans.merchant_category_id))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_hist_trans.merchant_id))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_hist_trans.category_2))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_hist_trans.state_id))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_hist_trans.subsector_id))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_hist_trans.purchased_on_weekend))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(AVG_TIME_BETWEEN(temp_hist_trans.purchase_date))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(TIME_SINCE_LAST(temp_hist_trans.purchase_date))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(TREND(temp_hist_trans.month_diff, purchase_date))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(TREND(temp_hist_trans.installments, purchase_date))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(TREND(temp_hist_trans.amount_month_ratio, purchase_date))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(TREND(temp_hist_trans.purchase_amount, purchase_date))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(TREND(temp_hist_trans.purchase_date_successive_diff, purchase_date))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(TREND(temp_hist_trans.duration, purchase_date))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(TREND(temp_hist_trans.month_lag, purchase_date))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(TREND(temp_hist_trans.purchased_on_weekday, purchase_date))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SUM(temp_new_trans.installments))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SUM(temp_new_trans.month_lag))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SUM(temp_new_trans.purchase_amount))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SUM(temp_new_trans.purchased_on_weekday))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SUM(temp_new_trans.month_diff))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SUM(temp_new_trans.purchase_date_successive_diff))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SUM(temp_new_trans.duration))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SUM(temp_new_trans.amount_month_ratio))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(STD(temp_new_trans.installments))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(STD(temp_new_trans.month_lag))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(STD(temp_new_trans.purchase_amount))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(STD(temp_new_trans.purchased_on_weekday))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(STD(temp_new_trans.month_diff))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(STD(temp_new_trans.purchase_date_successive_diff))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(STD(temp_new_trans.duration))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(STD(temp_new_trans.amount_month_ratio))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MAX(temp_new_trans.installments))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MAX(temp_new_trans.month_lag))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MAX(temp_new_trans.purchase_amount))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MAX(temp_new_trans.purchased_on_weekday))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MAX(temp_new_trans.month_diff))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MAX(temp_new_trans.purchase_date_successive_diff))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MAX(temp_new_trans.duration))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MAX(temp_new_trans.amount_month_ratio))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SKEW(temp_new_trans.installments))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SKEW(temp_new_trans.month_lag))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SKEW(temp_new_trans.purchase_amount))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SKEW(temp_new_trans.purchased_on_weekday))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SKEW(temp_new_trans.month_diff))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SKEW(temp_new_trans.purchase_date_successive_diff))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SKEW(temp_new_trans.duration))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(SKEW(temp_new_trans.amount_month_ratio))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MIN(temp_new_trans.installments))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MIN(temp_new_trans.month_lag))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MIN(temp_new_trans.purchase_amount))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MIN(temp_new_trans.purchased_on_weekday))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MIN(temp_new_trans.month_diff))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MIN(temp_new_trans.purchase_date_successive_diff))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MIN(temp_new_trans.duration))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MIN(temp_new_trans.amount_month_ratio))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEAN(temp_new_trans.installments))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEAN(temp_new_trans.month_lag))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEAN(temp_new_trans.purchase_amount))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEAN(temp_new_trans.purchased_on_weekday))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEAN(temp_new_trans.month_diff))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEAN(temp_new_trans.purchase_date_successive_diff))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEAN(temp_new_trans.duration))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEAN(temp_new_trans.amount_month_ratio))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(COUNT(temp_new_trans))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEDIAN(temp_new_trans.installments))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEDIAN(temp_new_trans.month_lag))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEDIAN(temp_new_trans.purchase_amount))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEDIAN(temp_new_trans.purchased_on_weekday))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEDIAN(temp_new_trans.month_diff))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEDIAN(temp_new_trans.purchase_date_successive_diff))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEDIAN(temp_new_trans.duration))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(MEDIAN(temp_new_trans.amount_month_ratio))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_new_trans.authorized_flag))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_new_trans.city_id))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_new_trans.category_1))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_new_trans.category_3))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_new_trans.merchant_category_id))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_new_trans.merchant_id))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_new_trans.category_2))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_new_trans.state_id))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_new_trans.subsector_id))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(NUM_UNIQUE(temp_new_trans.purchased_on_weekend))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(AVG_TIME_BETWEEN(temp_new_trans.purchase_date))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(TIME_SINCE_LAST(temp_new_trans.purchase_date))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(TREND(temp_new_trans.installments, purchase_date))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(TREND(temp_new_trans.amount_month_ratio, purchase_date))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(TREND(temp_new_trans.purchase_amount, purchase_date))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(TREND(temp_new_trans.purchase_date_successive_diff, purchase_date))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(TREND(temp_new_trans.duration, purchase_date))</th>\n",
       "      <td>1</td>\n",
       "      <td>0.5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(TREND(temp_new_trans.purchased_on_weekday, purchase_date))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(TREND(temp_new_trans.month_diff, purchase_date))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(TREND(temp_new_trans.month_lag, purchase_date))</th>\n",
       "      <td>0.5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>PERCENTILE(DAYS_SINCE(first_active_month))</th>\n",
       "      <td>0.75</td>\n",
       "      <td>0.75</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "card_id                                             C_ID_92a2005557  \\\n",
       "target                                                    -0.820283   \n",
       "feature_1                                                         5   \n",
       "feature_2                                                         2   \n",
       "feature_3                                                         1   \n",
       "SUM(temp_hist_trans.installments)                                 4   \n",
       "SUM(temp_hist_trans.month_lag)                                -1017   \n",
       "SUM(temp_hist_trans.purchase_amount)                       -165.969   \n",
       "SUM(temp_hist_trans.purchased_on_weekday)                       170   \n",
       "SUM(temp_hist_trans.month_diff)                                2951   \n",
       "SUM(temp_hist_trans.purchase_date_successive_diff)       2.0978e+07   \n",
       "SUM(temp_hist_trans.duration)                              -1885.49   \n",
       "SUM(temp_hist_trans.amount_month_ratio)                    -14.6348   \n",
       "STD(temp_hist_trans.installments)                          0.123314   \n",
       "STD(temp_hist_trans.month_lag)                              2.39769   \n",
       "STD(temp_hist_trans.purchase_amount)                       0.212139   \n",
       "STD(temp_hist_trans.purchased_on_weekday)                   0.47666   \n",
       "STD(temp_hist_trans.month_diff)                             0.47789   \n",
       "STD(temp_hist_trans.purchase_date_successive_diff)           107376   \n",
       "STD(temp_hist_trans.duration)                               2.39005   \n",
       "STD(temp_hist_trans.amount_month_ratio)                   0.0191685   \n",
       "MAX(temp_hist_trans.installments)                                 1   \n",
       "MAX(temp_hist_trans.month_lag)                                    0   \n",
       "MAX(temp_hist_trans.purchase_amount)                        2.25839   \n",
       "MAX(temp_hist_trans.purchased_on_weekday)                         1   \n",
       "MAX(temp_hist_trans.month_diff)                                  12   \n",
       "MAX(temp_hist_trans.purchase_date_successive_diff)      1.04474e+06   \n",
       "MAX(temp_hist_trans.duration)                               24.8423   \n",
       "MAX(temp_hist_trans.amount_month_ratio)                    0.205309   \n",
       "SKEW(temp_hist_trans.installments)                          7.92077   \n",
       "SKEW(temp_hist_trans.month_lag)                           0.0659603   \n",
       "SKEW(temp_hist_trans.purchase_amount)                       10.2431   \n",
       "SKEW(temp_hist_trans.purchased_on_weekday)                -0.650521   \n",
       "SKEW(temp_hist_trans.month_diff)                           0.632627   \n",
       "SKEW(temp_hist_trans.purchase_date_successive_d...          3.89701   \n",
       "SKEW(temp_hist_trans.duration)                              9.75999   \n",
       "SKEW(temp_hist_trans.amount_month_ratio)                    10.1906   \n",
       "MIN(temp_hist_trans.installments)                                 0   \n",
       "MIN(temp_hist_trans.month_lag)                                   -8   \n",
       "MIN(temp_hist_trans.purchase_amount)                      -0.739395   \n",
       "MIN(temp_hist_trans.purchased_on_weekday)                         0   \n",
       "MIN(temp_hist_trans.month_diff)                                  11   \n",
       "MIN(temp_hist_trans.purchase_date_successive_diff)                0   \n",
       "MIN(temp_hist_trans.duration)                              -8.80602   \n",
       "MIN(temp_hist_trans.amount_month_ratio)                  -0.0672177   \n",
       "MEAN(temp_hist_trans.installments)                        0.0153846   \n",
       "MEAN(temp_hist_trans.month_lag)                            -3.91154   \n",
       "MEAN(temp_hist_trans.purchase_amount)                     -0.638341   \n",
       "MEAN(temp_hist_trans.purchased_on_weekday)                 0.653846   \n",
       "MEAN(temp_hist_trans.month_diff)                              11.35   \n",
       "MEAN(temp_hist_trans.purchase_date_successive_d...          80684.6   \n",
       "MEAN(temp_hist_trans.duration)                             -7.25187   \n",
       "MEAN(temp_hist_trans.amount_month_ratio)                 -0.0562877   \n",
       "COUNT(temp_hist_trans)                                          260   \n",
       "MEDIAN(temp_hist_trans.installments)                              0   \n",
       "MEDIAN(temp_hist_trans.month_lag)                                -4   \n",
       "MEDIAN(temp_hist_trans.purchase_amount)                   -0.698042   \n",
       "MEDIAN(temp_hist_trans.purchased_on_weekday)                      1   \n",
       "MEDIAN(temp_hist_trans.month_diff)                               11   \n",
       "MEDIAN(temp_hist_trans.purchase_date_successive...          62157.5   \n",
       "MEDIAN(temp_hist_trans.duration)                            -7.8022   \n",
       "MEDIAN(temp_hist_trans.amount_month_ratio)               -0.0595747   \n",
       "NUM_UNIQUE(temp_hist_trans.authorized_flag)                       2   \n",
       "NUM_UNIQUE(temp_hist_trans.city_id)                               7   \n",
       "NUM_UNIQUE(temp_hist_trans.category_1)                            1   \n",
       "NUM_UNIQUE(temp_hist_trans.category_3)                            2   \n",
       "NUM_UNIQUE(temp_hist_trans.merchant_category_id)                 41   \n",
       "NUM_UNIQUE(temp_hist_trans.merchant_id)                          94   \n",
       "NUM_UNIQUE(temp_hist_trans.category_2)                            2   \n",
       "NUM_UNIQUE(temp_hist_trans.state_id)                              3   \n",
       "NUM_UNIQUE(temp_hist_trans.subsector_id)                         21   \n",
       "NUM_UNIQUE(temp_hist_trans.purchased_on_weekend)                  2   \n",
       "MODE(temp_hist_trans.authorized_flag)                             1   \n",
       "MODE(temp_hist_trans.city_id)                                    69   \n",
       "MODE(temp_hist_trans.category_1)                                  0   \n",
       "MODE(temp_hist_trans.category_3)                                  0   \n",
       "MODE(temp_hist_trans.merchant_category_id)                      560   \n",
       "MODE(temp_hist_trans.merchant_id)                   M_ID_1a81c358a3   \n",
       "MODE(temp_hist_trans.category_2)                                  1   \n",
       "MODE(temp_hist_trans.state_id)                                    9   \n",
       "MODE(temp_hist_trans.subsector_id)                               34   \n",
       "MODE(temp_hist_trans.purchased_on_weekend)                        0   \n",
       "AVG_TIME_BETWEEN(temp_hist_trans.purchase_date)             80996.1   \n",
       "TIME_SINCE_LAST(temp_hist_trans.purchase_date)          5.07859e+07   \n",
       "TREND(temp_hist_trans.month_diff, purchase_date)       -2.03099e-08   \n",
       "TREND(temp_hist_trans.installments, purchase_date)      -8.9272e-10   \n",
       "TREND(temp_hist_trans.amount_month_ratio, purch...      3.09112e-10   \n",
       "TREND(temp_hist_trans.purchase_amount, purchase...      4.55973e-09   \n",
       "TREND(temp_hist_trans.purchase_date_successive_...       0.00159676   \n",
       "TREND(temp_hist_trans.duration, purchase_date)           6.4071e-08   \n",
       "TREND(temp_hist_trans.month_lag, purchase_date)         3.71169e-07   \n",
       "TREND(temp_hist_trans.purchased_on_weekday, pur...     -6.85407e-09   \n",
       "SUM(temp_new_trans.installments)                                  0   \n",
       "SUM(temp_new_trans.month_lag)                                    34   \n",
       "SUM(temp_new_trans.purchase_amount)                        -13.2442   \n",
       "SUM(temp_new_trans.purchased_on_weekday)                         17   \n",
       "SUM(temp_new_trans.month_diff)                                  261   \n",
       "SUM(temp_new_trans.purchase_date_successive_diff)       4.74231e+06   \n",
       "SUM(temp_new_trans.duration)                               -150.358   \n",
       "SUM(temp_new_trans.amount_month_ratio)                     -1.16863   \n",
       "STD(temp_new_trans.installments)                                  0   \n",
       "STD(temp_new_trans.month_lag)                              0.510754   \n",
       "STD(temp_new_trans.purchase_amount)                        0.135812   \n",
       "STD(temp_new_trans.purchased_on_weekday)                   0.448978   \n",
       "STD(temp_new_trans.month_diff)                             0.486985   \n",
       "STD(temp_new_trans.purchase_date_successive_diff)            205878   \n",
       "STD(temp_new_trans.duration)                                1.57293   \n",
       "STD(temp_new_trans.amount_month_ratio)                    0.0121497   \n",
       "MAX(temp_new_trans.installments)                                  0   \n",
       "MAX(temp_new_trans.month_lag)                                     2   \n",
       "MAX(temp_new_trans.purchase_amount)                       -0.296112   \n",
       "MAX(temp_new_trans.purchased_on_weekday)                          1   \n",
       "MAX(temp_new_trans.month_diff)                                   12   \n",
       "MAX(temp_new_trans.purchase_date_successive_diff)            778975   \n",
       "MAX(temp_new_trans.duration)                               -3.25724   \n",
       "MAX(temp_new_trans.amount_month_ratio)                   -0.0256465   \n",
       "SKEW(temp_new_trans.installments)                                 0   \n",
       "SKEW(temp_new_trans.month_lag)                            0.0932329   \n",
       "SKEW(temp_new_trans.purchase_amount)                       0.899623   \n",
       "SKEW(temp_new_trans.purchased_on_weekday)                  -1.16667   \n",
       "SKEW(temp_new_trans.month_diff)                            0.684484   \n",
       "SKEW(temp_new_trans.purchase_date_successive_diff)          1.69853   \n",
       "SKEW(temp_new_trans.duration)                              0.857694   \n",
       "SKEW(temp_new_trans.amount_month_ratio)                    0.771362   \n",
       "MIN(temp_new_trans.installments)                                  0   \n",
       "MIN(temp_new_trans.month_lag)                                     1   \n",
       "MIN(temp_new_trans.purchase_amount)                       -0.724368   \n",
       "MIN(temp_new_trans.purchased_on_weekday)                          0   \n",
       "MIN(temp_new_trans.month_diff)                                   11   \n",
       "MIN(temp_new_trans.purchase_date_successive_diff)                 0   \n",
       "MIN(temp_new_trans.duration)                               -8.63868   \n",
       "MIN(temp_new_trans.amount_month_ratio)                   -0.0658516   \n",
       "MEAN(temp_new_trans.installments)                                 0   \n",
       "MEAN(temp_new_trans.month_lag)                              1.47826   \n",
       "MEAN(temp_new_trans.purchase_amount)                      -0.575835   \n",
       "MEAN(temp_new_trans.purchased_on_weekday)                   0.73913   \n",
       "MEAN(temp_new_trans.month_diff)                             11.3478   \n",
       "MEAN(temp_new_trans.purchase_date_successive_diff)           206187   \n",
       "MEAN(temp_new_trans.duration)                              -6.53729   \n",
       "MEAN(temp_new_trans.amount_month_ratio)                  -0.0508099   \n",
       "COUNT(temp_new_trans)                                            23   \n",
       "MEDIAN(temp_new_trans.installments)                               0   \n",
       "MEDIAN(temp_new_trans.month_lag)                                  1   \n",
       "MEDIAN(temp_new_trans.purchase_amount)                     -0.58118   \n",
       "MEDIAN(temp_new_trans.purchased_on_weekday)                       1   \n",
       "MEDIAN(temp_new_trans.month_diff)                                11   \n",
       "MEDIAN(temp_new_trans.purchase_date_successive_...           164976   \n",
       "MEDIAN(temp_new_trans.duration)                            -6.83875   \n",
       "MEDIAN(temp_new_trans.amount_month_ratio)                -0.0488853   \n",
       "NUM_UNIQUE(temp_new_trans.authorized_flag)                        1   \n",
       "NUM_UNIQUE(temp_new_trans.city_id)                                3   \n",
       "NUM_UNIQUE(temp_new_trans.category_1)                             1   \n",
       "NUM_UNIQUE(temp_new_trans.category_3)                             1   \n",
       "NUM_UNIQUE(temp_new_trans.merchant_category_id)                  14   \n",
       "NUM_UNIQUE(temp_new_trans.merchant_id)                           23   \n",
       "NUM_UNIQUE(temp_new_trans.category_2)                             1   \n",
       "NUM_UNIQUE(temp_new_trans.state_id)                               1   \n",
       "NUM_UNIQUE(temp_new_trans.subsector_id)                          10   \n",
       "NUM_UNIQUE(temp_new_trans.purchased_on_weekend)                   2   \n",
       "MODE(temp_new_trans.authorized_flag)                              1   \n",
       "MODE(temp_new_trans.city_id)                                     69   \n",
       "MODE(temp_new_trans.category_1)                                   0   \n",
       "MODE(temp_new_trans.category_3)                                   0   \n",
       "MODE(temp_new_trans.merchant_category_id)                       278   \n",
       "MODE(temp_new_trans.merchant_id)                    M_ID_0703c10cc0   \n",
       "MODE(temp_new_trans.category_2)                                   1   \n",
       "MODE(temp_new_trans.state_id)                                     9   \n",
       "MODE(temp_new_trans.subsector_id)                                37   \n",
       "MODE(temp_new_trans.purchased_on_weekend)                         0   \n",
       "AVG_TIME_BETWEEN(temp_new_trans.purchase_date)               215560   \n",
       "TIME_SINCE_LAST(temp_new_trans.purchase_date)           2.91004e+07   \n",
       "TREND(temp_new_trans.installments, purchase_date)                 0   \n",
       "TREND(temp_new_trans.amount_month_ratio, purcha...     -1.46783e-09   \n",
       "TREND(temp_new_trans.purchase_amount, purchase_...     -1.10757e-08   \n",
       "TREND(temp_new_trans.purchase_date_successive_d...          0.03212   \n",
       "TREND(temp_new_trans.duration, purchase_date)          -6.09874e-08   \n",
       "TREND(temp_new_trans.purchased_on_weekday, purc...     -5.64721e-08   \n",
       "TREND(temp_new_trans.month_diff, purchase_date)        -8.02519e-08   \n",
       "TREND(temp_new_trans.month_lag, purchase_date)          2.96858e-07   \n",
       "DAY(first_active_month)                                           1   \n",
       "MONTH(first_active_month)                                         6   \n",
       "WEEKDAY(first_active_month)                                       3   \n",
       "DAYS_SINCE(first_active_month)                                  614   \n",
       "PERCENTILE(target)                                              0.5   \n",
       "HOUR(first_active_month)                                          0   \n",
       "SUM(temp_hist_trans.installments WHERE purchase...                1   \n",
       "SUM(temp_hist_trans.installments WHERE authoriz...                0   \n",
       "SUM(temp_hist_trans.month_lag WHERE purchased_o...             -326   \n",
       "SUM(temp_hist_trans.month_lag WHERE authorized_...             -959   \n",
       "SUM(temp_hist_trans.purchase_amount WHERE purch...         -55.8855   \n",
       "SUM(temp_hist_trans.purchase_amount WHERE autho...         -157.397   \n",
       "SUM(temp_hist_trans.purchased_on_weekday WHERE ...                0   \n",
       "SUM(temp_hist_trans.purchased_on_weekday WHERE ...              161   \n",
       "SUM(temp_hist_trans.month_diff WHERE purchased_...             1017   \n",
       "SUM(temp_hist_trans.month_diff WHERE authorized...             2802   \n",
       "SUM(temp_hist_trans.purchase_date_successive_di...      7.96363e+06   \n",
       "SUM(temp_hist_trans.purchase_date_successive_di...      1.94592e+07   \n",
       "SUM(temp_hist_trans.duration WHERE purchased_on...         -630.755   \n",
       "SUM(temp_hist_trans.duration WHERE authorized_f...         -1787.31   \n",
       "SUM(temp_hist_trans.amount_month_ratio WHERE pu...         -4.95917   \n",
       "SUM(temp_hist_trans.amount_month_ratio WHERE au...          -13.885   \n",
       "SUM(temp_hist_trans.DAYS_SINCE(purchase_date))               123197   \n",
       "SUM(temp_hist_trans.DAYS_SINCE(purchase_date) W...            41802   \n",
       "SUM(temp_hist_trans.DAYS_SINCE(purchase_date) W...           116792   \n",
       "SUM(temp_hist_trans.PERCENTILE(installments))               129.684   \n",
       "SUM(temp_hist_trans.PERCENTILE(installments) WH...          44.6985   \n",
       "SUM(temp_hist_trans.PERCENTILE(installments) WH...            121.3   \n",
       "SUM(temp_hist_trans.PERCENTILE(month_lag))                  113.419   \n",
       "SUM(temp_hist_trans.PERCENTILE(month_lag) WHERE...          42.1234   \n",
       "SUM(temp_hist_trans.PERCENTILE(month_lag) WHERE...          108.859   \n",
       "SUM(temp_hist_trans.PERCENTILE(purchase_amount))            130.022   \n",
       "SUM(temp_hist_trans.PERCENTILE(purchase_amount)...          51.9046   \n",
       "SUM(temp_hist_trans.PERCENTILE(purchase_amount)...          124.229   \n",
       "SUM(temp_hist_trans.PERCENTILE(purchased_on_wee...           131.31   \n",
       "SUM(temp_hist_trans.PERCENTILE(purchased_on_wee...          16.0305   \n",
       "SUM(temp_hist_trans.PERCENTILE(purchased_on_wee...          124.495   \n",
       "SUM(temp_hist_trans.PERCENTILE(month_diff))                 86.3359   \n",
       "SUM(temp_hist_trans.PERCENTILE(month_diff) WHER...          28.3969   \n",
       "SUM(temp_hist_trans.PERCENTILE(month_diff) WHER...          81.5394   \n",
       "SUM(temp_hist_trans.PERCENTILE(purchase_date_su...          135.628   \n",
       "SUM(temp_hist_trans.PERCENTILE(purchase_date_su...          46.6476   \n",
       "SUM(temp_hist_trans.PERCENTILE(purchase_date_su...          129.425   \n",
       "SUM(temp_hist_trans.PERCENTILE(duration))                   158.537   \n",
       "SUM(temp_hist_trans.PERCENTILE(duration) WHERE ...          61.1387   \n",
       "SUM(temp_hist_trans.PERCENTILE(duration) WHERE ...          151.294   \n",
       "SUM(temp_hist_trans.PERCENTILE(amount_month_rat...          96.2774   \n",
       "SUM(temp_hist_trans.PERCENTILE(amount_month_rat...          38.3702   \n",
       "SUM(temp_hist_trans.PERCENTILE(amount_month_rat...          91.6578   \n",
       "STD(temp_hist_trans.DAYS_SINCE(purchase_date))              74.2452   \n",
       "STD(temp_hist_trans.PERCENTILE(installments))             0.0616571   \n",
       "STD(temp_hist_trans.PERCENTILE(month_lag))                   0.2978   \n",
       "STD(temp_hist_trans.PERCENTILE(purchase_amount))           0.273598   \n",
       "STD(temp_hist_trans.PERCENTILE(purchased_on_wee...          0.23833   \n",
       "STD(temp_hist_trans.PERCENTILE(month_diff))                 0.15808   \n",
       "STD(temp_hist_trans.PERCENTILE(purchase_date_su...         0.280444   \n",
       "STD(temp_hist_trans.PERCENTILE(duration))                   0.20574   \n",
       "STD(temp_hist_trans.PERCENTILE(amount_month_rat...         0.258088   \n",
       "MAX(temp_hist_trans.DAYS_SINCE(purchase_date))                  587   \n",
       "MAX(temp_hist_trans.PERCENTILE(installments))              0.991094   \n",
       "MAX(temp_hist_trans.PERCENTILE(month_lag))                 0.959288   \n",
       "MAX(temp_hist_trans.PERCENTILE(purchase_amount))           0.994911   \n",
       "MAX(temp_hist_trans.PERCENTILE(purchased_on_wee...         0.678117   \n",
       "MAX(temp_hist_trans.PERCENTILE(month_diff))                0.547074   \n",
       "MAX(temp_hist_trans.PERCENTILE(purchase_date_su...         0.992366   \n",
       "MAX(temp_hist_trans.PERCENTILE(duration))                  0.994911   \n",
       "MAX(temp_hist_trans.PERCENTILE(amount_month_rat...                1   \n",
       "SKEW(temp_hist_trans.DAYS_SINCE(purchase_date))          -0.0534497   \n",
       "SKEW(temp_hist_trans.PERCENTILE(installments))              7.92077   \n",
       "SKEW(temp_hist_trans.PERCENTILE(month_lag))                0.290522   \n",
       "SKEW(temp_hist_trans.PERCENTILE(purchase_amount))         0.0065437   \n",
       "SKEW(temp_hist_trans.PERCENTILE(purchased_on_we...        -0.650521   \n",
       "SKEW(temp_hist_trans.PERCENTILE(month_diff))               0.632627   \n",
       "SKEW(temp_hist_trans.PERCENTILE(purchase_date_s...        -0.158507   \n",
       "SKEW(temp_hist_trans.PERCENTILE(duration))               0.00399592   \n",
       "SKEW(temp_hist_trans.PERCENTILE(amount_month_ra...         0.745192   \n",
       "MIN(temp_hist_trans.DAYS_SINCE(purchase_date))                  344   \n",
       "MIN(temp_hist_trans.PERCENTILE(installments))              0.491094   \n",
       "MIN(temp_hist_trans.PERCENTILE(month_lag))               0.00508906   \n",
       "MIN(temp_hist_trans.PERCENTILE(purchase_amount))          0.0114504   \n",
       "MIN(temp_hist_trans.PERCENTILE(purchased_on_wee...         0.178117   \n",
       "MIN(temp_hist_trans.PERCENTILE(month_diff))                0.216285   \n",
       "MIN(temp_hist_trans.PERCENTILE(purchase_date_su...       0.00508906   \n",
       "MIN(temp_hist_trans.PERCENTILE(duration))                  0.246819   \n",
       "MIN(temp_hist_trans.PERCENTILE(amount_month_rat...       0.00254453   \n",
       "MEAN(temp_hist_trans.installments WHERE purchas...        0.0111111   \n",
       "MEAN(temp_hist_trans.installments WHERE authori...                0   \n",
       "MEAN(temp_hist_trans.month_lag WHERE purchased_...         -3.62222   \n",
       "MEAN(temp_hist_trans.month_lag WHERE authorized...         -3.88259   \n",
       "MEAN(temp_hist_trans.purchase_amount WHERE purc...         -0.62095   \n",
       "MEAN(temp_hist_trans.purchase_amount WHERE auth...        -0.637235   \n",
       "MEAN(temp_hist_trans.purchased_on_weekday WHERE...                0   \n",
       "MEAN(temp_hist_trans.purchased_on_weekday WHERE...         0.651822   \n",
       "MEAN(temp_hist_trans.month_diff WHERE purchased...             11.3   \n",
       "MEAN(temp_hist_trans.month_diff WHERE authorize...          11.3441   \n",
       "MEAN(temp_hist_trans.purchase_date_successive_d...          88484.8   \n",
       "MEAN(temp_hist_trans.purchase_date_successive_d...          78782.3   \n",
       "MEAN(temp_hist_trans.duration WHERE purchased_o...         -7.00839   \n",
       "MEAN(temp_hist_trans.duration WHERE authorized_...         -7.23609   \n",
       "MEAN(temp_hist_trans.amount_month_ratio WHERE p...       -0.0551019   \n",
       "MEAN(temp_hist_trans.amount_month_ratio WHERE a...       -0.0562145   \n",
       "MEAN(temp_hist_trans.DAYS_SINCE(purchase_date))             473.835   \n",
       "MEAN(temp_hist_trans.DAYS_SINCE(purchase_date) ...          464.467   \n",
       "MEAN(temp_hist_trans.DAYS_SINCE(purchase_date) ...          472.842   \n",
       "MEAN(temp_hist_trans.PERCENTILE(installments))             0.498786   \n",
       "MEAN(temp_hist_trans.PERCENTILE(installments) W...          0.49665   \n",
       "MEAN(temp_hist_trans.PERCENTILE(installments) W...         0.491094   \n",
       "MEAN(temp_hist_trans.PERCENTILE(month_lag))                0.436225   \n",
       "MEAN(temp_hist_trans.PERCENTILE(month_lag) WHER...         0.468038   \n",
       "MEAN(temp_hist_trans.PERCENTILE(month_lag) WHER...         0.440724   \n",
       "MEAN(temp_hist_trans.PERCENTILE(purchase_amount))          0.500083   \n",
       "MEAN(temp_hist_trans.PERCENTILE(purchase_amount...         0.576718   \n",
       "MEAN(temp_hist_trans.PERCENTILE(purchase_amount...         0.502951   \n",
       "MEAN(temp_hist_trans.PERCENTILE(purchased_on_we...          0.50504   \n",
       "MEAN(temp_hist_trans.PERCENTILE(purchased_on_we...         0.178117   \n",
       "MEAN(temp_hist_trans.PERCENTILE(purchased_on_we...         0.504028   \n",
       "MEAN(temp_hist_trans.PERCENTILE(month_diff))               0.332061   \n",
       "MEAN(temp_hist_trans.PERCENTILE(month_diff) WHE...         0.315522   \n",
       "MEAN(temp_hist_trans.PERCENTILE(month_diff) WHE...         0.330119   \n",
       "MEAN(temp_hist_trans.PERCENTILE(purchase_date_s...         0.521648   \n",
       "MEAN(temp_hist_trans.PERCENTILE(purchase_date_s...         0.518306   \n",
       "MEAN(temp_hist_trans.PERCENTILE(purchase_date_s...         0.523988   \n",
       "MEAN(temp_hist_trans.PERCENTILE(duration))                 0.609757   \n",
       "MEAN(temp_hist_trans.PERCENTILE(duration) WHERE...         0.679319   \n",
       "MEAN(temp_hist_trans.PERCENTILE(duration) WHERE...         0.612526   \n",
       "MEAN(temp_hist_trans.PERCENTILE(amount_month_ra...         0.370298   \n",
       "MEAN(temp_hist_trans.PERCENTILE(amount_month_ra...         0.426336   \n",
       "MEAN(temp_hist_trans.PERCENTILE(amount_month_ra...         0.371084   \n",
       "MEDIAN(temp_hist_trans.DAYS_SINCE(purchase_date))             475.5   \n",
       "MEDIAN(temp_hist_trans.PERCENTILE(installments))           0.491094   \n",
       "MEDIAN(temp_hist_trans.PERCENTILE(month_lag))              0.382952   \n",
       "MEDIAN(temp_hist_trans.PERCENTILE(purchase_amou...         0.510178   \n",
       "MEDIAN(temp_hist_trans.PERCENTILE(purchased_on_...         0.678117   \n",
       "MEDIAN(temp_hist_trans.PERCENTILE(month_diff))             0.216285   \n",
       "MEDIAN(temp_hist_trans.PERCENTILE(purchase_date...         0.553435   \n",
       "MEDIAN(temp_hist_trans.PERCENTILE(duration))                0.60687   \n",
       "MEDIAN(temp_hist_trans.PERCENTILE(amount_month_...         0.332061   \n",
       "NUM_UNIQUE(temp_hist_trans.DAY(purchase_date))                   31   \n",
       "NUM_UNIQUE(temp_hist_trans.MONTH(purchase_date))                  9   \n",
       "NUM_UNIQUE(temp_hist_trans.WEEKDAY(purchase_date))                7   \n",
       "NUM_UNIQUE(temp_hist_trans.HOUR(purchase_date))                  23   \n",
       "MODE(temp_hist_trans.DAY(purchase_date))                         11   \n",
       "MODE(temp_hist_trans.MONTH(purchase_date))                       12   \n",
       "MODE(temp_hist_trans.WEEKDAY(purchase_date))                      5   \n",
       "MODE(temp_hist_trans.HOUR(purchase_date))                        14   \n",
       "TREND(temp_hist_trans.PERCENTILE(month_lag), pu...      4.58304e-08   \n",
       "TREND(temp_hist_trans.PERCENTILE(duration), pur...       8.4611e-09   \n",
       "TREND(temp_hist_trans.PERCENTILE(amount_month_r...      4.19797e-09   \n",
       "TREND(temp_hist_trans.PERCENTILE(purchase_amoun...      6.24958e-09   \n",
       "TREND(temp_hist_trans.DAYS_SINCE(purchase_date)...     -1.15739e-05   \n",
       "TREND(temp_hist_trans.PERCENTILE(installments),...      -4.4636e-10   \n",
       "TREND(temp_hist_trans.PERCENTILE(purchased_on_w...     -3.42703e-09   \n",
       "TREND(temp_hist_trans.PERCENTILE(purchase_date_...      2.49699e-09   \n",
       "TREND(temp_hist_trans.PERCENTILE(month_diff), p...     -6.71828e-09   \n",
       "SUM(temp_new_trans.installments WHERE purchased...                0   \n",
       "SUM(temp_new_trans.month_lag WHERE purchased_on...                9   \n",
       "SUM(temp_new_trans.purchase_amount WHERE purcha...         -2.81409   \n",
       "SUM(temp_new_trans.purchased_on_weekday WHERE p...                0   \n",
       "SUM(temp_new_trans.month_diff WHERE purchased_o...               67   \n",
       "SUM(temp_new_trans.purchase_date_successive_dif...      1.48362e+06   \n",
       "SUM(temp_new_trans.duration WHERE purchased_on_...         -31.2627   \n",
       "SUM(temp_new_trans.amount_month_ratio WHERE pur...        -0.253495   \n",
       "SUM(temp_new_trans.DAYS_SINCE(purchase_date))                  7125   \n",
       "SUM(temp_new_trans.DAYS_SINCE(purchase_date) WH...             1829   \n",
       "SUM(temp_new_trans.PERCENTILE(installments))                11.7447   \n",
       "SUM(temp_new_trans.PERCENTILE(installments) WHE...          3.06383   \n",
       "SUM(temp_new_trans.PERCENTILE(month_lag))                   12.5957   \n",
       "SUM(temp_new_trans.PERCENTILE(month_lag) WHERE ...          3.35106   \n",
       "SUM(temp_new_trans.PERCENTILE(purchase_amount))             11.6277   \n",
       "SUM(temp_new_trans.PERCENTILE(purchase_amount) ...          4.28723   \n",
       "SUM(temp_new_trans.PERCENTILE(purchased_on_week...          13.1489   \n",
       "SUM(temp_new_trans.PERCENTILE(purchased_on_week...          1.21277   \n",
       "SUM(temp_new_trans.PERCENTILE(month_diff))                  5.87234   \n",
       "SUM(temp_new_trans.PERCENTILE(month_diff) WHERE...          1.26596   \n",
       "SUM(temp_new_trans.PERCENTILE(purchase_date_suc...          12.3511   \n",
       "SUM(temp_new_trans.PERCENTILE(purchase_date_suc...          3.68085   \n",
       "SUM(temp_new_trans.PERCENTILE(duration))                    14.4043   \n",
       "SUM(temp_new_trans.PERCENTILE(duration) WHERE p...          4.71277   \n",
       "SUM(temp_new_trans.PERCENTILE(amount_month_ratio))          8.80851   \n",
       "SUM(temp_new_trans.PERCENTILE(amount_month_rati...          3.62766   \n",
       "STD(temp_new_trans.DAYS_SINCE(purchase_date))               16.6022   \n",
       "STD(temp_new_trans.PERCENTILE(installments))                      0   \n",
       "STD(temp_new_trans.PERCENTILE(month_lag))                  0.255377   \n",
       "STD(temp_new_trans.PERCENTILE(purchase_amount))            0.271089   \n",
       "STD(temp_new_trans.PERCENTILE(purchased_on_week...         0.224489   \n",
       "STD(temp_new_trans.PERCENTILE(month_diff))                 0.119156   \n",
       "STD(temp_new_trans.PERCENTILE(purchase_date_suc...          0.26951   \n",
       "STD(temp_new_trans.PERCENTILE(duration))                   0.199755   \n",
       "STD(temp_new_trans.PERCENTILE(amount_month_ratio))         0.286876   \n",
       "MAX(temp_new_trans.DAYS_SINCE(purchase_date))                   336   \n",
       "MAX(temp_new_trans.PERCENTILE(installments))               0.510638   \n",
       "MAX(temp_new_trans.PERCENTILE(month_lag))                  0.808511   \n",
       "MAX(temp_new_trans.PERCENTILE(purchase_amount))            0.904255   \n",
       "MAX(temp_new_trans.PERCENTILE(purchased_on_week...         0.702128   \n",
       "MAX(temp_new_trans.PERCENTILE(month_diff))                 0.414894   \n",
       "MAX(temp_new_trans.PERCENTILE(purchase_date_suc...                1   \n",
       "MAX(temp_new_trans.PERCENTILE(duration))                   0.925532   \n",
       "MAX(temp_new_trans.PERCENTILE(amount_month_ratio))         0.914894   \n",
       "SKEW(temp_new_trans.DAYS_SINCE(purchase_date))            0.0263921   \n",
       "SKEW(temp_new_trans.PERCENTILE(installments))                     0   \n",
       "SKEW(temp_new_trans.PERCENTILE(month_lag))                0.0932329   \n",
       "SKEW(temp_new_trans.PERCENTILE(purchase_amount))          -0.113279   \n",
       "SKEW(temp_new_trans.PERCENTILE(purchased_on_wee...         -1.16667   \n",
       "SKEW(temp_new_trans.PERCENTILE(month_diff))                0.684484   \n",
       "SKEW(temp_new_trans.PERCENTILE(purchase_date_su...        -0.127311   \n",
       "SKEW(temp_new_trans.PERCENTILE(duration))                 -0.128077   \n",
       "SKEW(temp_new_trans.PERCENTILE(amount_month_rat...          0.50927   \n",
       "MIN(temp_new_trans.DAYS_SINCE(purchase_date))                   281   \n",
       "MIN(temp_new_trans.PERCENTILE(installments))               0.510638   \n",
       "MIN(temp_new_trans.PERCENTILE(month_lag))                  0.308511   \n",
       "MIN(temp_new_trans.PERCENTILE(purchase_amount))           0.0638298   \n",
       "MIN(temp_new_trans.PERCENTILE(purchased_on_week...         0.202128   \n",
       "MIN(temp_new_trans.PERCENTILE(month_diff))                 0.170213   \n",
       "MIN(temp_new_trans.PERCENTILE(purchase_date_suc...        0.0319149   \n",
       "MIN(temp_new_trans.PERCENTILE(duration))                   0.276596   \n",
       "MIN(temp_new_trans.PERCENTILE(amount_month_ratio))        0.0212766   \n",
       "MEAN(temp_new_trans.installments WHERE purchase...                0   \n",
       "MEAN(temp_new_trans.month_lag WHERE purchased_o...              1.5   \n",
       "MEAN(temp_new_trans.purchase_amount WHERE purch...        -0.469015   \n",
       "MEAN(temp_new_trans.purchased_on_weekday WHERE ...                0   \n",
       "MEAN(temp_new_trans.month_diff WHERE purchased_...          11.1667   \n",
       "MEAN(temp_new_trans.purchase_date_successive_di...           247271   \n",
       "MEAN(temp_new_trans.duration WHERE purchased_on...         -5.21046   \n",
       "MEAN(temp_new_trans.amount_month_ratio WHERE pu...       -0.0422491   \n",
       "MEAN(temp_new_trans.DAYS_SINCE(purchase_date))              309.783   \n",
       "MEAN(temp_new_trans.DAYS_SINCE(purchase_date) W...          304.833   \n",
       "MEAN(temp_new_trans.PERCENTILE(installments))              0.510638   \n",
       "MEAN(temp_new_trans.PERCENTILE(installments) WH...         0.510638   \n",
       "MEAN(temp_new_trans.PERCENTILE(month_lag))                 0.547641   \n",
       "MEAN(temp_new_trans.PERCENTILE(month_lag) WHERE...         0.558511   \n",
       "MEAN(temp_new_trans.PERCENTILE(purchase_amount))            0.50555   \n",
       "MEAN(temp_new_trans.PERCENTILE(purchase_amount)...         0.714539   \n",
       "MEAN(temp_new_trans.PERCENTILE(purchased_on_wee...         0.571693   \n",
       "MEAN(temp_new_trans.PERCENTILE(purchased_on_wee...         0.202128   \n",
       "MEAN(temp_new_trans.PERCENTILE(month_diff))                0.255319   \n",
       "MEAN(temp_new_trans.PERCENTILE(month_diff) WHER...         0.210993   \n",
       "MEAN(temp_new_trans.PERCENTILE(purchase_date_su...         0.537003   \n",
       "MEAN(temp_new_trans.PERCENTILE(purchase_date_su...         0.613475   \n",
       "MEAN(temp_new_trans.PERCENTILE(duration))                  0.626272   \n",
       "MEAN(temp_new_trans.PERCENTILE(duration) WHERE ...         0.785461   \n",
       "MEAN(temp_new_trans.PERCENTILE(amount_month_rat...         0.382979   \n",
       "MEAN(temp_new_trans.PERCENTILE(amount_month_rat...          0.60461   \n",
       "MEDIAN(temp_new_trans.DAYS_SINCE(purchase_date))                310   \n",
       "MEDIAN(temp_new_trans.PERCENTILE(installments))            0.510638   \n",
       "MEDIAN(temp_new_trans.PERCENTILE(month_lag))               0.308511   \n",
       "MEDIAN(temp_new_trans.PERCENTILE(purchase_amount))         0.553191   \n",
       "MEDIAN(temp_new_trans.PERCENTILE(purchased_on_w...         0.702128   \n",
       "MEDIAN(temp_new_trans.PERCENTILE(month_diff))              0.170213   \n",
       "MEDIAN(temp_new_trans.PERCENTILE(purchase_date_...         0.553191   \n",
       "MEDIAN(temp_new_trans.PERCENTILE(duration))                0.638298   \n",
       "MEDIAN(temp_new_trans.PERCENTILE(amount_month_r...         0.404255   \n",
       "NUM_UNIQUE(temp_new_trans.DAY(purchase_date))                    17   \n",
       "NUM_UNIQUE(temp_new_trans.MONTH(purchase_date))                   2   \n",
       "NUM_UNIQUE(temp_new_trans.WEEKDAY(purchase_date))                 7   \n",
       "NUM_UNIQUE(temp_new_trans.HOUR(purchase_date))                    8   \n",
       "MODE(temp_new_trans.DAY(purchase_date))                           6   \n",
       "MODE(temp_new_trans.MONTH(purchase_date))                         3   \n",
       "MODE(temp_new_trans.WEEKDAY(purchase_date))                       4   \n",
       "MODE(temp_new_trans.HOUR(purchase_date))                         13   \n",
       "TREND(temp_new_trans.PERCENTILE(purchased_on_we...      -2.8236e-08   \n",
       "TREND(temp_new_trans.PERCENTILE(duration), purc...      1.50261e-09   \n",
       "TREND(temp_new_trans.PERCENTILE(month_diff), pu...     -1.96361e-08   \n",
       "TREND(temp_new_trans.PERCENTILE(amount_month_ra...     -2.25804e-08   \n",
       "TREND(temp_new_trans.PERCENTILE(installments), ...                0   \n",
       "TREND(temp_new_trans.DAYS_SINCE(purchase_date),...     -1.15877e-05   \n",
       "TREND(temp_new_trans.PERCENTILE(purchase_amount...     -1.18799e-08   \n",
       "TREND(temp_new_trans.PERCENTILE(month_lag), pur...      1.48429e-07   \n",
       "TREND(temp_new_trans.PERCENTILE(purchase_date_s...       3.6465e-08   \n",
       "PERCENTILE(SUM(temp_hist_trans.installments))                  0.75   \n",
       "PERCENTILE(SUM(temp_hist_trans.month_lag))                      0.5   \n",
       "PERCENTILE(SUM(temp_hist_trans.purchase_amount))                0.5   \n",
       "PERCENTILE(SUM(temp_hist_trans.purchased_on_wee...                1   \n",
       "PERCENTILE(SUM(temp_hist_trans.month_diff))                       1   \n",
       "PERCENTILE(SUM(temp_hist_trans.purchase_date_su...                1   \n",
       "PERCENTILE(SUM(temp_hist_trans.duration))                       0.5   \n",
       "PERCENTILE(SUM(temp_hist_trans.amount_month_rat...              0.5   \n",
       "PERCENTILE(STD(temp_hist_trans.installments))                   0.5   \n",
       "PERCENTILE(STD(temp_hist_trans.month_lag))                        1   \n",
       "PERCENTILE(STD(temp_hist_trans.purchase_amount))                0.5   \n",
       "PERCENTILE(STD(temp_hist_trans.purchased_on_wee...              0.5   \n",
       "PERCENTILE(STD(temp_hist_trans.month_diff))                     0.5   \n",
       "PERCENTILE(STD(temp_hist_trans.purchase_date_su...              0.5   \n",
       "PERCENTILE(STD(temp_hist_trans.duration))                       0.5   \n",
       "PERCENTILE(STD(temp_hist_trans.amount_month_rat...              0.5   \n",
       "PERCENTILE(MAX(temp_hist_trans.installments))                  0.75   \n",
       "PERCENTILE(MAX(temp_hist_trans.month_lag))                     0.75   \n",
       "PERCENTILE(MAX(temp_hist_trans.purchase_amount))                0.5   \n",
       "PERCENTILE(MAX(temp_hist_trans.purchased_on_wee...             0.75   \n",
       "PERCENTILE(MAX(temp_hist_trans.month_diff))                     0.5   \n",
       "PERCENTILE(MAX(temp_hist_trans.purchase_date_su...              0.5   \n",
       "PERCENTILE(MAX(temp_hist_trans.duration))                       0.5   \n",
       "PERCENTILE(MAX(temp_hist_trans.amount_month_rat...                1   \n",
       "PERCENTILE(SKEW(temp_hist_trans.installments))                    1   \n",
       "PERCENTILE(SKEW(temp_hist_trans.month_lag))                       1   \n",
       "PERCENTILE(SKEW(temp_hist_trans.purchase_amount))                 1   \n",
       "PERCENTILE(SKEW(temp_hist_trans.purchased_on_we...              0.5   \n",
       "PERCENTILE(SKEW(temp_hist_trans.month_diff))                      1   \n",
       "PERCENTILE(SKEW(temp_hist_trans.purchase_date_s...              0.5   \n",
       "PERCENTILE(SKEW(temp_hist_trans.duration))                        1   \n",
       "PERCENTILE(SKEW(temp_hist_trans.amount_month_ra...                1   \n",
       "PERCENTILE(MIN(temp_hist_trans.installments))                  0.75   \n",
       "PERCENTILE(MIN(temp_hist_trans.month_lag))                      0.5   \n",
       "PERCENTILE(MIN(temp_hist_trans.purchase_amount))                  1   \n",
       "PERCENTILE(MIN(temp_hist_trans.purchased_on_wee...             0.75   \n",
       "PERCENTILE(MIN(temp_hist_trans.month_diff))                     0.5   \n",
       "PERCENTILE(MIN(temp_hist_trans.purchase_date_su...             0.75   \n",
       "PERCENTILE(MIN(temp_hist_trans.duration))                         1   \n",
       "PERCENTILE(MIN(temp_hist_trans.amount_month_rat...              0.5   \n",
       "PERCENTILE(MEAN(temp_hist_trans.installments))                  0.5   \n",
       "PERCENTILE(MEAN(temp_hist_trans.month_lag))                     0.5   \n",
       "PERCENTILE(MEAN(temp_hist_trans.purchase_amount))               0.5   \n",
       "PERCENTILE(MEAN(temp_hist_trans.purchased_on_we...                1   \n",
       "PERCENTILE(MEAN(temp_hist_trans.month_diff))                    0.5   \n",
       "PERCENTILE(MEAN(temp_hist_trans.purchase_date_s...              0.5   \n",
       "PERCENTILE(MEAN(temp_hist_trans.duration))                        1   \n",
       "PERCENTILE(MEAN(temp_hist_trans.amount_month_ra...              0.5   \n",
       "PERCENTILE(COUNT(temp_hist_trans))                                1   \n",
       "PERCENTILE(MEDIAN(temp_hist_trans.installments))               0.75   \n",
       "PERCENTILE(MEDIAN(temp_hist_trans.month_lag))                   0.5   \n",
       "PERCENTILE(MEDIAN(temp_hist_trans.purchase_amou...                1   \n",
       "PERCENTILE(MEDIAN(temp_hist_trans.purchased_on_...             0.75   \n",
       "PERCENTILE(MEDIAN(temp_hist_trans.month_diff))                  0.5   \n",
       "PERCENTILE(MEDIAN(temp_hist_trans.purchase_date...                1   \n",
       "PERCENTILE(MEDIAN(temp_hist_trans.duration))                      1   \n",
       "PERCENTILE(MEDIAN(temp_hist_trans.amount_month_...              0.5   \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.authorize...             0.75   \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.city_id))                   1   \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.category_1))             0.75   \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.category_3))             0.75   \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.merchant_...                1   \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.merchant_...                1   \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.category_2))             0.75   \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.state_id))                  1   \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.subsector...                1   \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.purchased...             0.75   \n",
       "PERCENTILE(AVG_TIME_BETWEEN(temp_hist_trans.pur...              0.5   \n",
       "PERCENTILE(TIME_SINCE_LAST(temp_hist_trans.purc...              0.5   \n",
       "PERCENTILE(TREND(temp_hist_trans.month_diff, pu...                1   \n",
       "PERCENTILE(TREND(temp_hist_trans.installments, ...              0.5   \n",
       "PERCENTILE(TREND(temp_hist_trans.amount_month_r...                1   \n",
       "PERCENTILE(TREND(temp_hist_trans.purchase_amoun...                1   \n",
       "PERCENTILE(TREND(temp_hist_trans.purchase_date_...              0.5   \n",
       "PERCENTILE(TREND(temp_hist_trans.duration, purc...                1   \n",
       "PERCENTILE(TREND(temp_hist_trans.month_lag, pur...                1   \n",
       "PERCENTILE(TREND(temp_hist_trans.purchased_on_w...              0.5   \n",
       "PERCENTILE(SUM(temp_new_trans.installments))                   0.75   \n",
       "PERCENTILE(SUM(temp_new_trans.month_lag))                         1   \n",
       "PERCENTILE(SUM(temp_new_trans.purchase_amount))                 0.5   \n",
       "PERCENTILE(SUM(temp_new_trans.purchased_on_week...                1   \n",
       "PERCENTILE(SUM(temp_new_trans.month_diff))                      0.5   \n",
       "PERCENTILE(SUM(temp_new_trans.purchase_date_suc...                1   \n",
       "PERCENTILE(SUM(temp_new_trans.duration))                        0.5   \n",
       "PERCENTILE(SUM(temp_new_trans.amount_month_ratio))              0.5   \n",
       "PERCENTILE(STD(temp_new_trans.installments))                   0.75   \n",
       "PERCENTILE(STD(temp_new_trans.month_lag))                         1   \n",
       "PERCENTILE(STD(temp_new_trans.purchase_amount))                 0.5   \n",
       "PERCENTILE(STD(temp_new_trans.purchased_on_week...              0.5   \n",
       "PERCENTILE(STD(temp_new_trans.month_diff))                        1   \n",
       "PERCENTILE(STD(temp_new_trans.purchase_date_suc...              0.5   \n",
       "PERCENTILE(STD(temp_new_trans.duration))                        0.5   \n",
       "PERCENTILE(STD(temp_new_trans.amount_month_ratio))              0.5   \n",
       "PERCENTILE(MAX(temp_new_trans.installments))                   0.75   \n",
       "PERCENTILE(MAX(temp_new_trans.month_lag))                      0.75   \n",
       "PERCENTILE(MAX(temp_new_trans.purchase_amount))                 0.5   \n",
       "PERCENTILE(MAX(temp_new_trans.purchased_on_week...             0.75   \n",
       "PERCENTILE(MAX(temp_new_trans.month_diff))                      0.5   \n",
       "PERCENTILE(MAX(temp_new_trans.purchase_date_suc...                1   \n",
       "PERCENTILE(MAX(temp_new_trans.duration))                        0.5   \n",
       "PERCENTILE(MAX(temp_new_trans.amount_month_ratio))              0.5   \n",
       "PERCENTILE(SKEW(temp_new_trans.installments))                  0.75   \n",
       "PERCENTILE(SKEW(temp_new_trans.month_lag))                      0.5   \n",
       "PERCENTILE(SKEW(temp_new_trans.purchase_amount))                0.5   \n",
       "PERCENTILE(SKEW(temp_new_trans.purchased_on_wee...              0.5   \n",
       "PERCENTILE(SKEW(temp_new_trans.month_diff))                     0.5   \n",
       "PERCENTILE(SKEW(temp_new_trans.purchase_date_su...                1   \n",
       "PERCENTILE(SKEW(temp_new_trans.duration))                       0.5   \n",
       "PERCENTILE(SKEW(temp_new_trans.amount_month_rat...              0.5   \n",
       "PERCENTILE(MIN(temp_new_trans.installments))                   0.75   \n",
       "PERCENTILE(MIN(temp_new_trans.month_lag))                      0.75   \n",
       "PERCENTILE(MIN(temp_new_trans.purchase_amount))                   1   \n",
       "PERCENTILE(MIN(temp_new_trans.purchased_on_week...             0.75   \n",
       "PERCENTILE(MIN(temp_new_trans.month_diff))                      0.5   \n",
       "PERCENTILE(MIN(temp_new_trans.purchase_date_suc...             0.75   \n",
       "PERCENTILE(MIN(temp_new_trans.duration))                          1   \n",
       "PERCENTILE(MIN(temp_new_trans.amount_month_ratio))              0.5   \n",
       "PERCENTILE(MEAN(temp_new_trans.installments))                  0.75   \n",
       "PERCENTILE(MEAN(temp_new_trans.month_lag))                        1   \n",
       "PERCENTILE(MEAN(temp_new_trans.purchase_amount))                0.5   \n",
       "PERCENTILE(MEAN(temp_new_trans.purchased_on_wee...                1   \n",
       "PERCENTILE(MEAN(temp_new_trans.month_diff))                     0.5   \n",
       "PERCENTILE(MEAN(temp_new_trans.purchase_date_su...                1   \n",
       "PERCENTILE(MEAN(temp_new_trans.duration))                       0.5   \n",
       "PERCENTILE(MEAN(temp_new_trans.amount_month_rat...              0.5   \n",
       "PERCENTILE(COUNT(temp_new_trans))                               0.5   \n",
       "PERCENTILE(MEDIAN(temp_new_trans.installments))                0.75   \n",
       "PERCENTILE(MEDIAN(temp_new_trans.month_lag))                   0.75   \n",
       "PERCENTILE(MEDIAN(temp_new_trans.purchase_amount))                1   \n",
       "PERCENTILE(MEDIAN(temp_new_trans.purchased_on_w...                1   \n",
       "PERCENTILE(MEDIAN(temp_new_trans.month_diff))                   0.5   \n",
       "PERCENTILE(MEDIAN(temp_new_trans.purchase_date_...                1   \n",
       "PERCENTILE(MEDIAN(temp_new_trans.duration))                       1   \n",
       "PERCENTILE(MEDIAN(temp_new_trans.amount_month_r...              0.5   \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.authorized...             0.75   \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.city_id))                  0.5   \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.category_1))              0.75   \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.category_3))              0.75   \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.merchant_c...             0.75   \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.merchant_id))              0.5   \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.category_2))               0.5   \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.state_id))                 0.5   \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.subsector_...              0.5   \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.purchased_...             0.75   \n",
       "PERCENTILE(AVG_TIME_BETWEEN(temp_new_trans.purc...                1   \n",
       "PERCENTILE(TIME_SINCE_LAST(temp_new_trans.purch...              0.5   \n",
       "PERCENTILE(TREND(temp_new_trans.installments, p...             0.75   \n",
       "PERCENTILE(TREND(temp_new_trans.amount_month_ra...                1   \n",
       "PERCENTILE(TREND(temp_new_trans.purchase_amount...                1   \n",
       "PERCENTILE(TREND(temp_new_trans.purchase_date_s...              0.5   \n",
       "PERCENTILE(TREND(temp_new_trans.duration, purch...                1   \n",
       "PERCENTILE(TREND(temp_new_trans.purchased_on_we...              0.5   \n",
       "PERCENTILE(TREND(temp_new_trans.month_diff, pur...              0.5   \n",
       "PERCENTILE(TREND(temp_new_trans.month_lag, purc...              0.5   \n",
       "PERCENTILE(DAYS_SINCE(first_active_month))                     0.75   \n",
       "\n",
       "card_id                                             C_ID_ae86f7d5fb  \n",
       "target                                                    -0.563362  \n",
       "feature_1                                                         5  \n",
       "feature_2                                                         2  \n",
       "feature_3                                                         1  \n",
       "SUM(temp_hist_trans.installments)                                 4  \n",
       "SUM(temp_hist_trans.month_lag)                                 -308  \n",
       "SUM(temp_hist_trans.purchase_amount)                       -74.8775  \n",
       "SUM(temp_hist_trans.purchased_on_weekday)                        84  \n",
       "SUM(temp_hist_trans.month_diff)                                1933  \n",
       "SUM(temp_hist_trans.purchase_date_successive_diff)      1.45427e+07  \n",
       "SUM(temp_hist_trans.duration)                              -1085.41  \n",
       "SUM(temp_hist_trans.amount_month_ratio)                    -5.17162  \n",
       "STD(temp_hist_trans.installments)                           0.17144  \n",
       "STD(temp_hist_trans.month_lag)                              1.40005  \n",
       "STD(temp_hist_trans.purchase_amount)                       0.415112  \n",
       "STD(temp_hist_trans.purchased_on_weekday)                    0.4842  \n",
       "STD(temp_hist_trans.month_diff)                             0.50074  \n",
       "STD(temp_hist_trans.purchase_date_successive_diff)           246098  \n",
       "STD(temp_hist_trans.duration)                               6.13736  \n",
       "STD(temp_hist_trans.amount_month_ratio)                   0.0281872  \n",
       "MAX(temp_hist_trans.installments)                                 1  \n",
       "MAX(temp_hist_trans.month_lag)                                    0  \n",
       "MAX(temp_hist_trans.purchase_amount)                         2.2716  \n",
       "MAX(temp_hist_trans.purchased_on_weekday)                         1  \n",
       "MAX(temp_hist_trans.month_diff)                                  15  \n",
       "MAX(temp_hist_trans.purchase_date_successive_diff)      1.73702e+06  \n",
       "MAX(temp_hist_trans.duration)                                34.074  \n",
       "MAX(temp_hist_trans.amount_month_ratio)                     0.15144  \n",
       "SKEW(temp_hist_trans.installments)                          5.56579  \n",
       "SKEW(temp_hist_trans.month_lag)                           -0.158893  \n",
       "SKEW(temp_hist_trans.purchase_amount)                       5.23007  \n",
       "SKEW(temp_hist_trans.purchased_on_weekday)                -0.551787  \n",
       "SKEW(temp_hist_trans.month_diff)                          -0.137202  \n",
       "SKEW(temp_hist_trans.purchase_date_successive_d...          4.40674  \n",
       "SKEW(temp_hist_trans.duration)                              5.31526  \n",
       "SKEW(temp_hist_trans.amount_month_ratio)                    5.09217  \n",
       "MIN(temp_hist_trans.installments)                                 0  \n",
       "MIN(temp_hist_trans.month_lag)                                   -5  \n",
       "MIN(temp_hist_trans.purchase_amount)                      -0.740942  \n",
       "MIN(temp_hist_trans.purchased_on_weekday)                         0  \n",
       "MIN(temp_hist_trans.month_diff)                                  14  \n",
       "MIN(temp_hist_trans.purchase_date_successive_diff)                0  \n",
       "MIN(temp_hist_trans.duration)                              -11.0909  \n",
       "MIN(temp_hist_trans.amount_month_ratio)                  -0.0529244  \n",
       "MEAN(temp_hist_trans.installments)                        0.0300752  \n",
       "MEAN(temp_hist_trans.month_lag)                            -2.31579  \n",
       "MEAN(temp_hist_trans.purchase_amount)                     -0.562989  \n",
       "MEAN(temp_hist_trans.purchased_on_weekday)                 0.631579  \n",
       "MEAN(temp_hist_trans.month_diff)                            14.5338  \n",
       "MEAN(temp_hist_trans.purchase_date_successive_d...           109344  \n",
       "MEAN(temp_hist_trans.duration)                             -8.16095  \n",
       "MEAN(temp_hist_trans.amount_month_ratio)                 -0.0388844  \n",
       "COUNT(temp_hist_trans)                                          133  \n",
       "MEDIAN(temp_hist_trans.installments)                              0  \n",
       "MEDIAN(temp_hist_trans.month_lag)                                -2  \n",
       "MEDIAN(temp_hist_trans.purchase_amount)                   -0.701828  \n",
       "MEDIAN(temp_hist_trans.purchased_on_weekday)                      1  \n",
       "MEDIAN(temp_hist_trans.month_diff)                               15  \n",
       "MEDIAN(temp_hist_trans.purchase_date_successive...            29646  \n",
       "MEDIAN(temp_hist_trans.duration)                           -9.95182  \n",
       "MEDIAN(temp_hist_trans.amount_month_ratio)               -0.0469106  \n",
       "NUM_UNIQUE(temp_hist_trans.authorized_flag)                       2  \n",
       "NUM_UNIQUE(temp_hist_trans.city_id)                               3  \n",
       "NUM_UNIQUE(temp_hist_trans.category_1)                            1  \n",
       "NUM_UNIQUE(temp_hist_trans.category_3)                            2  \n",
       "NUM_UNIQUE(temp_hist_trans.merchant_category_id)                 27  \n",
       "NUM_UNIQUE(temp_hist_trans.merchant_id)                          74  \n",
       "NUM_UNIQUE(temp_hist_trans.category_2)                            2  \n",
       "NUM_UNIQUE(temp_hist_trans.state_id)                              2  \n",
       "NUM_UNIQUE(temp_hist_trans.subsector_id)                         18  \n",
       "NUM_UNIQUE(temp_hist_trans.purchased_on_weekend)                  2  \n",
       "MODE(temp_hist_trans.authorized_flag)                             1  \n",
       "MODE(temp_hist_trans.city_id)                                    87  \n",
       "MODE(temp_hist_trans.category_1)                                  0  \n",
       "MODE(temp_hist_trans.category_3)                                  0  \n",
       "MODE(temp_hist_trans.merchant_category_id)                      307  \n",
       "MODE(temp_hist_trans.merchant_id)                   M_ID_c8f44ec011  \n",
       "MODE(temp_hist_trans.category_2)                                  3  \n",
       "MODE(temp_hist_trans.state_id)                                   11  \n",
       "MODE(temp_hist_trans.subsector_id)                               37  \n",
       "MODE(temp_hist_trans.purchased_on_weekend)                        0  \n",
       "AVG_TIME_BETWEEN(temp_hist_trans.purchase_date)              110172  \n",
       "TIME_SINCE_LAST(temp_hist_trans.purchase_date)          5.19882e+07  \n",
       "TREND(temp_hist_trans.month_diff, purchase_date)       -5.28998e-08  \n",
       "TREND(temp_hist_trans.installments, purchase_date)     -3.35243e-10  \n",
       "TREND(temp_hist_trans.amount_month_ratio, purch...     -1.90272e-09  \n",
       "TREND(temp_hist_trans.purchase_amount, purchase...     -2.61004e-08  \n",
       "TREND(temp_hist_trans.purchase_date_successive_...       0.00385725  \n",
       "TREND(temp_hist_trans.duration, purchase_date)         -3.57342e-07  \n",
       "TREND(temp_hist_trans.month_lag, purchase_date)         3.42225e-07  \n",
       "TREND(temp_hist_trans.purchased_on_weekday, pur...      1.11143e-08  \n",
       "SUM(temp_new_trans.installments)                                  0  \n",
       "SUM(temp_new_trans.month_lag)                                    32  \n",
       "SUM(temp_new_trans.purchase_amount)                        -8.18339  \n",
       "SUM(temp_new_trans.purchased_on_weekday)                         12  \n",
       "SUM(temp_new_trans.month_diff)                                  342  \n",
       "SUM(temp_new_trans.purchase_date_successive_diff)       4.72866e+06  \n",
       "SUM(temp_new_trans.duration)                               -116.269  \n",
       "SUM(temp_new_trans.amount_month_ratio)                    -0.576425  \n",
       "STD(temp_new_trans.installments)                                  0  \n",
       "STD(temp_new_trans.month_lag)                              0.481543  \n",
       "STD(temp_new_trans.purchase_amount)                        0.734419  \n",
       "STD(temp_new_trans.purchased_on_weekday)                   0.510754  \n",
       "STD(temp_new_trans.month_diff)                             0.442326  \n",
       "STD(temp_new_trans.purchase_date_successive_diff)            222107  \n",
       "STD(temp_new_trans.duration)                                10.4416  \n",
       "STD(temp_new_trans.amount_month_ratio)                    0.0517468  \n",
       "MAX(temp_new_trans.installments)                                  0  \n",
       "MAX(temp_new_trans.month_lag)                                     2  \n",
       "MAX(temp_new_trans.purchase_amount)                         2.07808  \n",
       "MAX(temp_new_trans.purchased_on_weekday)                          1  \n",
       "MAX(temp_new_trans.month_diff)                                   15  \n",
       "MAX(temp_new_trans.purchase_date_successive_diff)            764226  \n",
       "MAX(temp_new_trans.duration)                                29.0931  \n",
       "MAX(temp_new_trans.amount_month_ratio)                     0.148434  \n",
       "SKEW(temp_new_trans.installments)                                 0  \n",
       "SKEW(temp_new_trans.month_lag)                             0.755148  \n",
       "SKEW(temp_new_trans.purchase_amount)                        2.51195  \n",
       "SKEW(temp_new_trans.purchased_on_weekday)                         0  \n",
       "SKEW(temp_new_trans.month_diff)                             1.23315  \n",
       "SKEW(temp_new_trans.purchase_date_successive_diff)          1.28946  \n",
       "SKEW(temp_new_trans.duration)                                2.4893  \n",
       "SKEW(temp_new_trans.amount_month_ratio)                      2.5319  \n",
       "MIN(temp_new_trans.installments)                                  0  \n",
       "MIN(temp_new_trans.month_lag)                                     1  \n",
       "MIN(temp_new_trans.purchase_amount)                       -0.727373  \n",
       "MIN(temp_new_trans.purchased_on_weekday)                          0  \n",
       "MIN(temp_new_trans.month_diff)                                   14  \n",
       "MIN(temp_new_trans.purchase_date_successive_diff)                 0  \n",
       "MIN(temp_new_trans.duration)                               -10.5725  \n",
       "MIN(temp_new_trans.amount_month_ratio)                   -0.0519552  \n",
       "MEAN(temp_new_trans.installments)                                 0  \n",
       "MEAN(temp_new_trans.month_lag)                              1.33333  \n",
       "MEAN(temp_new_trans.purchase_amount)                      -0.340975  \n",
       "MEAN(temp_new_trans.purchased_on_weekday)                       0.5  \n",
       "MEAN(temp_new_trans.month_diff)                               14.25  \n",
       "MEAN(temp_new_trans.purchase_date_successive_diff)           197027  \n",
       "MEAN(temp_new_trans.duration)                              -4.84455  \n",
       "MEAN(temp_new_trans.amount_month_ratio)                  -0.0240177  \n",
       "COUNT(temp_new_trans)                                            24  \n",
       "MEDIAN(temp_new_trans.installments)                               0  \n",
       "MEDIAN(temp_new_trans.month_lag)                                  1  \n",
       "MEDIAN(temp_new_trans.purchase_amount)                    -0.609438  \n",
       "MEDIAN(temp_new_trans.purchased_on_weekday)                     0.5  \n",
       "MEDIAN(temp_new_trans.month_diff)                                14  \n",
       "MEDIAN(temp_new_trans.purchase_date_successive_...           119832  \n",
       "MEDIAN(temp_new_trans.duration)                            -8.53213  \n",
       "MEDIAN(temp_new_trans.amount_month_ratio)                -0.0435313  \n",
       "NUM_UNIQUE(temp_new_trans.authorized_flag)                        1  \n",
       "NUM_UNIQUE(temp_new_trans.city_id)                                4  \n",
       "NUM_UNIQUE(temp_new_trans.category_1)                             1  \n",
       "NUM_UNIQUE(temp_new_trans.category_3)                             1  \n",
       "NUM_UNIQUE(temp_new_trans.merchant_category_id)                  14  \n",
       "NUM_UNIQUE(temp_new_trans.merchant_id)                           24  \n",
       "NUM_UNIQUE(temp_new_trans.category_2)                             2  \n",
       "NUM_UNIQUE(temp_new_trans.state_id)                               3  \n",
       "NUM_UNIQUE(temp_new_trans.subsector_id)                          12  \n",
       "NUM_UNIQUE(temp_new_trans.purchased_on_weekend)                   2  \n",
       "MODE(temp_new_trans.authorized_flag)                              1  \n",
       "MODE(temp_new_trans.city_id)                                     87  \n",
       "MODE(temp_new_trans.category_1)                                   0  \n",
       "MODE(temp_new_trans.category_3)                                   0  \n",
       "MODE(temp_new_trans.merchant_category_id)                       278  \n",
       "MODE(temp_new_trans.merchant_id)                    M_ID_00a6ca8a8a  \n",
       "MODE(temp_new_trans.category_2)                                   3  \n",
       "MODE(temp_new_trans.state_id)                                    11  \n",
       "MODE(temp_new_trans.subsector_id)                                37  \n",
       "MODE(temp_new_trans.purchased_on_weekend)                         0  \n",
       "AVG_TIME_BETWEEN(temp_new_trans.purchase_date)               205594  \n",
       "TIME_SINCE_LAST(temp_new_trans.purchase_date)           3.71267e+07  \n",
       "TREND(temp_new_trans.installments, purchase_date)                 0  \n",
       "TREND(temp_new_trans.amount_month_ratio, purcha...     -1.94178e-09  \n",
       "TREND(temp_new_trans.purchase_amount, purchase_...     -3.03324e-08  \n",
       "TREND(temp_new_trans.purchase_date_successive_d...        0.0588533  \n",
       "TREND(temp_new_trans.duration, purchase_date)          -4.71864e-07  \n",
       "TREND(temp_new_trans.purchased_on_weekday, purc...      1.27364e-07  \n",
       "TREND(temp_new_trans.month_diff, purchase_date)        -7.76171e-08  \n",
       "TREND(temp_new_trans.month_lag, purchase_date)          3.27168e-07  \n",
       "DAY(first_active_month)                                           1  \n",
       "MONTH(first_active_month)                                         6  \n",
       "WEEKDAY(first_active_month)                                       3  \n",
       "DAYS_SINCE(first_active_month)                                  614  \n",
       "PERCENTILE(target)                                                1  \n",
       "HOUR(first_active_month)                                          0  \n",
       "SUM(temp_hist_trans.installments WHERE purchase...                1  \n",
       "SUM(temp_hist_trans.installments WHERE authoriz...                0  \n",
       "SUM(temp_hist_trans.month_lag WHERE purchased_o...             -122  \n",
       "SUM(temp_hist_trans.month_lag WHERE authorized_...             -263  \n",
       "SUM(temp_hist_trans.purchase_amount WHERE purch...         -27.0708  \n",
       "SUM(temp_hist_trans.purchase_amount WHERE autho...         -69.2762  \n",
       "SUM(temp_hist_trans.purchased_on_weekday WHERE ...                0  \n",
       "SUM(temp_hist_trans.purchased_on_weekday WHERE ...               70  \n",
       "SUM(temp_hist_trans.month_diff WHERE purchased_...              713  \n",
       "SUM(temp_hist_trans.month_diff WHERE authorized...             1674  \n",
       "SUM(temp_hist_trans.purchase_date_successive_di...       3.9765e+06  \n",
       "SUM(temp_hist_trans.purchase_date_successive_di...      1.27357e+07  \n",
       "SUM(temp_hist_trans.duration WHERE purchased_on...         -392.683  \n",
       "SUM(temp_hist_trans.duration WHERE authorized_f...         -1007.95  \n",
       "SUM(temp_hist_trans.amount_month_ratio WHERE pu...         -1.86843  \n",
       "SUM(temp_hist_trans.amount_month_ratio WHERE au...         -4.76694  \n",
       "SUM(temp_hist_trans.DAYS_SINCE(purchase_date))                69124  \n",
       "SUM(temp_hist_trans.DAYS_SINCE(purchase_date) W...            25744  \n",
       "SUM(temp_hist_trans.DAYS_SINCE(purchase_date) W...            59684  \n",
       "SUM(temp_hist_trans.PERCENTILE(installments))               67.3155  \n",
       "SUM(temp_hist_trans.PERCENTILE(installments) WH...          24.5636  \n",
       "SUM(temp_hist_trans.PERCENTILE(installments) WH...          56.4758  \n",
       "SUM(temp_hist_trans.PERCENTILE(month_lag))                  83.5814  \n",
       "SUM(temp_hist_trans.PERCENTILE(month_lag) WHERE...           29.458  \n",
       "SUM(temp_hist_trans.PERCENTILE(month_lag) WHERE...          72.7519  \n",
       "SUM(temp_hist_trans.PERCENTILE(purchase_amount))            66.9784  \n",
       "SUM(temp_hist_trans.PERCENTILE(purchase_amount)...          29.1539  \n",
       "SUM(temp_hist_trans.PERCENTILE(purchase_amount)...          59.6221  \n",
       "SUM(temp_hist_trans.PERCENTILE(purchased_on_wee...          65.6896  \n",
       "SUM(temp_hist_trans.PERCENTILE(purchased_on_wee...          8.72774  \n",
       "SUM(temp_hist_trans.PERCENTILE(purchased_on_wee...          55.4835  \n",
       "SUM(temp_hist_trans.PERCENTILE(month_diff))                 110.664  \n",
       "SUM(temp_hist_trans.PERCENTILE(month_diff) WHER...          40.9135  \n",
       "SUM(temp_hist_trans.PERCENTILE(month_diff) WHER...          96.1285  \n",
       "SUM(temp_hist_trans.PERCENTILE(purchase_date_su...          61.3715  \n",
       "SUM(temp_hist_trans.PERCENTILE(purchase_date_su...          20.2608  \n",
       "SUM(temp_hist_trans.PERCENTILE(purchase_date_su...          52.5216  \n",
       "SUM(temp_hist_trans.PERCENTILE(duration))                   38.4631  \n",
       "SUM(temp_hist_trans.PERCENTILE(duration) WHERE ...          18.4924  \n",
       "SUM(temp_hist_trans.PERCENTILE(duration) WHERE ...          33.0534  \n",
       "SUM(temp_hist_trans.PERCENTILE(amount_month_rat...          100.723  \n",
       "SUM(temp_hist_trans.PERCENTILE(amount_month_rat...          38.8664  \n",
       "SUM(temp_hist_trans.PERCENTILE(amount_month_rat...          87.7735  \n",
       "STD(temp_hist_trans.DAYS_SINCE(purchase_date))              46.4585  \n",
       "STD(temp_hist_trans.PERCENTILE(installments))               0.08572  \n",
       "STD(temp_hist_trans.PERCENTILE(month_lag))                 0.208161  \n",
       "STD(temp_hist_trans.PERCENTILE(purchase_amount))           0.318093  \n",
       "STD(temp_hist_trans.PERCENTILE(purchased_on_wee...           0.2421  \n",
       "STD(temp_hist_trans.PERCENTILE(month_diff))               0.0847308  \n",
       "STD(temp_hist_trans.PERCENTILE(purchase_date_su...         0.302273  \n",
       "STD(temp_hist_trans.PERCENTILE(duration))                  0.310715  \n",
       "STD(temp_hist_trans.PERCENTILE(amount_month_rat...         0.132508  \n",
       "MAX(temp_hist_trans.DAYS_SINCE(purchase_date))                  601  \n",
       "MAX(temp_hist_trans.PERCENTILE(installments))              0.991094  \n",
       "MAX(temp_hist_trans.PERCENTILE(month_lag))                 0.959288  \n",
       "MAX(temp_hist_trans.PERCENTILE(purchase_amount))           0.998728  \n",
       "MAX(temp_hist_trans.PERCENTILE(purchased_on_wee...         0.678117  \n",
       "MAX(temp_hist_trans.PERCENTILE(month_diff))                0.910941  \n",
       "MAX(temp_hist_trans.PERCENTILE(purchase_date_su...                1  \n",
       "MAX(temp_hist_trans.PERCENTILE(duration))                  0.998728  \n",
       "MAX(temp_hist_trans.PERCENTILE(amount_month_rat...         0.996183  \n",
       "SKEW(temp_hist_trans.DAYS_SINCE(purchase_date))           0.0429153  \n",
       "SKEW(temp_hist_trans.PERCENTILE(installments))              5.56579  \n",
       "SKEW(temp_hist_trans.PERCENTILE(month_lag))              -0.0287405  \n",
       "SKEW(temp_hist_trans.PERCENTILE(purchase_amount))        -0.0135877  \n",
       "SKEW(temp_hist_trans.PERCENTILE(purchased_on_we...        -0.551787  \n",
       "SKEW(temp_hist_trans.PERCENTILE(month_diff))              -0.137202  \n",
       "SKEW(temp_hist_trans.PERCENTILE(purchase_date_s...         0.306568  \n",
       "SKEW(temp_hist_trans.PERCENTILE(duration))                  1.40217  \n",
       "SKEW(temp_hist_trans.PERCENTILE(amount_month_ra...         0.189281  \n",
       "MIN(temp_hist_trans.DAYS_SINCE(purchase_date))                  433  \n",
       "MIN(temp_hist_trans.PERCENTILE(installments))              0.491094  \n",
       "MIN(temp_hist_trans.PERCENTILE(month_lag))                 0.276081  \n",
       "MIN(temp_hist_trans.PERCENTILE(purchase_amount))         0.00381679  \n",
       "MIN(temp_hist_trans.PERCENTILE(purchased_on_wee...         0.178117  \n",
       "MIN(temp_hist_trans.PERCENTILE(month_diff))                 0.74173  \n",
       "MIN(temp_hist_trans.PERCENTILE(purchase_date_su...       0.00508906  \n",
       "MIN(temp_hist_trans.PERCENTILE(duration))                0.00254453  \n",
       "MIN(temp_hist_trans.PERCENTILE(amount_month_rat...         0.533079  \n",
       "MEAN(temp_hist_trans.installments WHERE purchas...        0.0204082  \n",
       "MEAN(temp_hist_trans.installments WHERE authori...                0  \n",
       "MEAN(temp_hist_trans.month_lag WHERE purchased_...          -2.4898  \n",
       "MEAN(temp_hist_trans.month_lag WHERE authorized...         -2.28696  \n",
       "MEAN(temp_hist_trans.purchase_amount WHERE purc...        -0.552466  \n",
       "MEAN(temp_hist_trans.purchase_amount WHERE auth...        -0.602401  \n",
       "MEAN(temp_hist_trans.purchased_on_weekday WHERE...                0  \n",
       "MEAN(temp_hist_trans.purchased_on_weekday WHERE...         0.608696  \n",
       "MEAN(temp_hist_trans.month_diff WHERE purchased...           14.551  \n",
       "MEAN(temp_hist_trans.month_diff WHERE authorize...          14.5565  \n",
       "MEAN(temp_hist_trans.purchase_date_successive_d...          81153.1  \n",
       "MEAN(temp_hist_trans.purchase_date_successive_d...           110745  \n",
       "MEAN(temp_hist_trans.duration WHERE purchased_o...         -8.01394  \n",
       "MEAN(temp_hist_trans.duration WHERE authorized_...          -8.7648  \n",
       "MEAN(temp_hist_trans.amount_month_ratio WHERE p...       -0.0381313  \n",
       "MEAN(temp_hist_trans.amount_month_ratio WHERE a...       -0.0414517  \n",
       "MEAN(temp_hist_trans.DAYS_SINCE(purchase_date))             519.729  \n",
       "MEAN(temp_hist_trans.DAYS_SINCE(purchase_date) ...          525.388  \n",
       "MEAN(temp_hist_trans.DAYS_SINCE(purchase_date) ...          518.991  \n",
       "MEAN(temp_hist_trans.PERCENTILE(installments))             0.506132  \n",
       "MEAN(temp_hist_trans.PERCENTILE(installments) W...         0.501298  \n",
       "MEAN(temp_hist_trans.PERCENTILE(installments) W...         0.491094  \n",
       "MEAN(temp_hist_trans.PERCENTILE(month_lag))                0.628432  \n",
       "MEAN(temp_hist_trans.PERCENTILE(month_lag) WHER...         0.601184  \n",
       "MEAN(temp_hist_trans.PERCENTILE(month_lag) WHER...         0.632625  \n",
       "MEAN(temp_hist_trans.PERCENTILE(purchase_amount))          0.503597  \n",
       "MEAN(temp_hist_trans.PERCENTILE(purchase_amount...         0.594978  \n",
       "MEAN(temp_hist_trans.PERCENTILE(purchase_amount...         0.518453  \n",
       "MEAN(temp_hist_trans.PERCENTILE(purchased_on_we...         0.493907  \n",
       "MEAN(temp_hist_trans.PERCENTILE(purchased_on_we...         0.178117  \n",
       "MEAN(temp_hist_trans.PERCENTILE(purchased_on_we...         0.482465  \n",
       "MEAN(temp_hist_trans.PERCENTILE(month_diff))               0.832061  \n",
       "MEAN(temp_hist_trans.PERCENTILE(month_diff) WHE...         0.834969  \n",
       "MEAN(temp_hist_trans.PERCENTILE(month_diff) WHE...           0.8359  \n",
       "MEAN(temp_hist_trans.PERCENTILE(purchase_date_s...          0.46144  \n",
       "MEAN(temp_hist_trans.PERCENTILE(purchase_date_s...         0.413486  \n",
       "MEAN(temp_hist_trans.PERCENTILE(purchase_date_s...          0.45671  \n",
       "MEAN(temp_hist_trans.PERCENTILE(duration))                 0.289196  \n",
       "MEAN(temp_hist_trans.PERCENTILE(duration) WHERE...         0.377395  \n",
       "MEAN(temp_hist_trans.PERCENTILE(duration) WHERE...         0.287421  \n",
       "MEAN(temp_hist_trans.PERCENTILE(amount_month_ra...         0.757313  \n",
       "MEAN(temp_hist_trans.PERCENTILE(amount_month_ra...         0.793192  \n",
       "MEAN(temp_hist_trans.PERCENTILE(amount_month_ra...         0.763248  \n",
       "MEDIAN(temp_hist_trans.DAYS_SINCE(purchase_date))               514  \n",
       "MEDIAN(temp_hist_trans.PERCENTILE(installments))           0.491094  \n",
       "MEDIAN(temp_hist_trans.PERCENTILE(month_lag))              0.655216  \n",
       "MEDIAN(temp_hist_trans.PERCENTILE(purchase_amou...         0.466921  \n",
       "MEDIAN(temp_hist_trans.PERCENTILE(purchased_on_...         0.678117  \n",
       "MEDIAN(temp_hist_trans.PERCENTILE(month_diff))             0.910941  \n",
       "MEDIAN(temp_hist_trans.PERCENTILE(purchase_date...         0.440204  \n",
       "MEDIAN(temp_hist_trans.PERCENTILE(duration))               0.170483  \n",
       "MEDIAN(temp_hist_trans.PERCENTILE(amount_month_...         0.745547  \n",
       "NUM_UNIQUE(temp_hist_trans.DAY(purchase_date))                   27  \n",
       "NUM_UNIQUE(temp_hist_trans.MONTH(purchase_date))                  6  \n",
       "NUM_UNIQUE(temp_hist_trans.WEEKDAY(purchase_date))                7  \n",
       "NUM_UNIQUE(temp_hist_trans.HOUR(purchase_date))                  22  \n",
       "MODE(temp_hist_trans.DAY(purchase_date))                         14  \n",
       "MODE(temp_hist_trans.MONTH(purchase_date))                        7  \n",
       "MODE(temp_hist_trans.WEEKDAY(purchase_date))                      5  \n",
       "MODE(temp_hist_trans.HOUR(purchase_date))                        20  \n",
       "TREND(temp_hist_trans.PERCENTILE(month_lag), pu...      5.09193e-08  \n",
       "TREND(temp_hist_trans.PERCENTILE(duration), pur...      -1.8733e-08  \n",
       "TREND(temp_hist_trans.PERCENTILE(amount_month_r...     -1.29852e-08  \n",
       "TREND(temp_hist_trans.PERCENTILE(purchase_amoun...     -2.40576e-08  \n",
       "TREND(temp_hist_trans.DAYS_SINCE(purchase_date)...     -1.15688e-05  \n",
       "TREND(temp_hist_trans.PERCENTILE(installments),...     -1.67622e-10  \n",
       "TREND(temp_hist_trans.PERCENTILE(purchased_on_w...      5.55714e-09  \n",
       "TREND(temp_hist_trans.PERCENTILE(purchase_date_...      9.76888e-09  \n",
       "TREND(temp_hist_trans.PERCENTILE(month_diff), p...     -8.95123e-09  \n",
       "SUM(temp_new_trans.installments WHERE purchased...                0  \n",
       "SUM(temp_new_trans.month_lag WHERE purchased_on...               14  \n",
       "SUM(temp_new_trans.purchase_amount WHERE purcha...         -4.59485  \n",
       "SUM(temp_new_trans.purchased_on_weekday WHERE p...                0  \n",
       "SUM(temp_new_trans.month_diff WHERE purchased_o...              171  \n",
       "SUM(temp_new_trans.purchase_date_successive_dif...      1.85167e+06  \n",
       "SUM(temp_new_trans.duration WHERE purchased_on_...         -65.8502  \n",
       "SUM(temp_new_trans.amount_month_ratio WHERE pur...        -0.320955  \n",
       "SUM(temp_new_trans.DAYS_SINCE(purchase_date))                  9703  \n",
       "SUM(temp_new_trans.DAYS_SINCE(purchase_date) WH...             4905  \n",
       "SUM(temp_new_trans.PERCENTILE(installments))                12.2553  \n",
       "SUM(temp_new_trans.PERCENTILE(installments) WHE...          6.12766  \n",
       "SUM(temp_new_trans.PERCENTILE(month_lag))                   11.4043  \n",
       "SUM(temp_new_trans.PERCENTILE(month_lag) WHERE ...          4.70213  \n",
       "SUM(temp_new_trans.PERCENTILE(purchase_amount))             12.3723  \n",
       "SUM(temp_new_trans.PERCENTILE(purchase_amount) ...          5.51064  \n",
       "SUM(temp_new_trans.PERCENTILE(purchased_on_week...          10.8511  \n",
       "SUM(temp_new_trans.PERCENTILE(purchased_on_week...          2.42553  \n",
       "SUM(temp_new_trans.PERCENTILE(month_diff))                  18.1277  \n",
       "SUM(temp_new_trans.PERCENTILE(month_diff) WHERE...          9.06383  \n",
       "SUM(temp_new_trans.PERCENTILE(purchase_date_suc...          11.6489  \n",
       "SUM(temp_new_trans.PERCENTILE(purchase_date_suc...          4.90426  \n",
       "SUM(temp_new_trans.PERCENTILE(duration))                    9.59574  \n",
       "SUM(temp_new_trans.PERCENTILE(duration) WHERE p...          4.34043  \n",
       "SUM(temp_new_trans.PERCENTILE(amount_month_ratio))          15.1915  \n",
       "SUM(temp_new_trans.PERCENTILE(amount_month_rati...          6.97872  \n",
       "STD(temp_new_trans.DAYS_SINCE(purchase_date))               14.5018  \n",
       "STD(temp_new_trans.PERCENTILE(installments))                      0  \n",
       "STD(temp_new_trans.PERCENTILE(month_lag))                  0.240772  \n",
       "STD(temp_new_trans.PERCENTILE(purchase_amount))             0.31591  \n",
       "STD(temp_new_trans.PERCENTILE(purchased_on_week...         0.255377  \n",
       "STD(temp_new_trans.PERCENTILE(month_diff))                 0.112934  \n",
       "STD(temp_new_trans.PERCENTILE(purchase_date_suc...         0.315221  \n",
       "STD(temp_new_trans.PERCENTILE(duration))                   0.325332  \n",
       "STD(temp_new_trans.PERCENTILE(amount_month_ratio))         0.244047  \n",
       "MAX(temp_new_trans.DAYS_SINCE(purchase_date))                   429  \n",
       "MAX(temp_new_trans.PERCENTILE(installments))               0.510638  \n",
       "MAX(temp_new_trans.PERCENTILE(month_lag))                  0.808511  \n",
       "MAX(temp_new_trans.PERCENTILE(purchase_amount))                   1  \n",
       "MAX(temp_new_trans.PERCENTILE(purchased_on_week...         0.702128  \n",
       "MAX(temp_new_trans.PERCENTILE(month_diff))                 0.946809  \n",
       "MAX(temp_new_trans.PERCENTILE(purchase_date_suc...         0.978723  \n",
       "MAX(temp_new_trans.PERCENTILE(duration))                          1  \n",
       "MAX(temp_new_trans.PERCENTILE(amount_month_ratio))                1  \n",
       "SKEW(temp_new_trans.DAYS_SINCE(purchase_date))            -0.217544  \n",
       "SKEW(temp_new_trans.PERCENTILE(installments))                     0  \n",
       "SKEW(temp_new_trans.PERCENTILE(month_lag))                 0.755148  \n",
       "SKEW(temp_new_trans.PERCENTILE(purchase_amount))          0.0557786  \n",
       "SKEW(temp_new_trans.PERCENTILE(purchased_on_wee...                0  \n",
       "SKEW(temp_new_trans.PERCENTILE(month_diff))                 1.23315  \n",
       "SKEW(temp_new_trans.PERCENTILE(purchase_date_su...         0.147175  \n",
       "SKEW(temp_new_trans.PERCENTILE(duration))                  0.719544  \n",
       "SKEW(temp_new_trans.PERCENTILE(amount_month_rat...        -0.153612  \n",
       "MIN(temp_new_trans.DAYS_SINCE(purchase_date))                   374  \n",
       "MIN(temp_new_trans.PERCENTILE(installments))               0.510638  \n",
       "MIN(temp_new_trans.PERCENTILE(month_lag))                  0.308511  \n",
       "MIN(temp_new_trans.PERCENTILE(purchase_amount))           0.0212766  \n",
       "MIN(temp_new_trans.PERCENTILE(purchased_on_week...         0.202128  \n",
       "MIN(temp_new_trans.PERCENTILE(month_diff))                 0.691489  \n",
       "MIN(temp_new_trans.PERCENTILE(purchase_date_suc...        0.0319149  \n",
       "MIN(temp_new_trans.PERCENTILE(duration))                  0.0212766  \n",
       "MIN(temp_new_trans.PERCENTILE(amount_month_ratio))         0.255319  \n",
       "MEAN(temp_new_trans.installments WHERE purchase...                0  \n",
       "MEAN(temp_new_trans.month_lag WHERE purchased_o...          1.16667  \n",
       "MEAN(temp_new_trans.purchase_amount WHERE purch...        -0.382904  \n",
       "MEAN(temp_new_trans.purchased_on_weekday WHERE ...                0  \n",
       "MEAN(temp_new_trans.month_diff WHERE purchased_...            14.25  \n",
       "MEAN(temp_new_trans.purchase_date_successive_di...           154306  \n",
       "MEAN(temp_new_trans.duration WHERE purchased_on...         -5.48752  \n",
       "MEAN(temp_new_trans.amount_month_ratio WHERE pu...       -0.0267462  \n",
       "MEAN(temp_new_trans.DAYS_SINCE(purchase_date))              404.292  \n",
       "MEAN(temp_new_trans.DAYS_SINCE(purchase_date) W...           408.75  \n",
       "MEAN(temp_new_trans.PERCENTILE(installments))              0.510638  \n",
       "MEAN(temp_new_trans.PERCENTILE(installments) WH...         0.510638  \n",
       "MEAN(temp_new_trans.PERCENTILE(month_lag))                 0.475177  \n",
       "MEAN(temp_new_trans.PERCENTILE(month_lag) WHERE...         0.391844  \n",
       "MEAN(temp_new_trans.PERCENTILE(purchase_amount))           0.515514  \n",
       "MEAN(temp_new_trans.PERCENTILE(purchase_amount)...          0.45922  \n",
       "MEAN(temp_new_trans.PERCENTILE(purchased_on_wee...         0.452128  \n",
       "MEAN(temp_new_trans.PERCENTILE(purchased_on_wee...         0.202128  \n",
       "MEAN(temp_new_trans.PERCENTILE(month_diff))                0.755319  \n",
       "MEAN(temp_new_trans.PERCENTILE(month_diff) WHER...         0.755319  \n",
       "MEAN(temp_new_trans.PERCENTILE(purchase_date_su...         0.485372  \n",
       "MEAN(temp_new_trans.PERCENTILE(purchase_date_su...         0.408688  \n",
       "MEAN(temp_new_trans.PERCENTILE(duration))                  0.399823  \n",
       "MEAN(temp_new_trans.PERCENTILE(duration) WHERE ...         0.361702  \n",
       "MEAN(temp_new_trans.PERCENTILE(amount_month_rat...         0.632979  \n",
       "MEAN(temp_new_trans.PERCENTILE(amount_month_rat...          0.58156  \n",
       "MEDIAN(temp_new_trans.DAYS_SINCE(purchase_date))                407  \n",
       "MEDIAN(temp_new_trans.PERCENTILE(installments))            0.510638  \n",
       "MEDIAN(temp_new_trans.PERCENTILE(month_lag))               0.308511  \n",
       "MEDIAN(temp_new_trans.PERCENTILE(purchase_amount))         0.484043  \n",
       "MEDIAN(temp_new_trans.PERCENTILE(purchased_on_w...         0.452128  \n",
       "MEDIAN(temp_new_trans.PERCENTILE(month_diff))              0.691489  \n",
       "MEDIAN(temp_new_trans.PERCENTILE(purchase_date_...         0.468085  \n",
       "MEDIAN(temp_new_trans.PERCENTILE(duration))                0.287234  \n",
       "MEDIAN(temp_new_trans.PERCENTILE(amount_month_r...         0.680851  \n",
       "NUM_UNIQUE(temp_new_trans.DAY(purchase_date))                    17  \n",
       "NUM_UNIQUE(temp_new_trans.MONTH(purchase_date))                   2  \n",
       "NUM_UNIQUE(temp_new_trans.WEEKDAY(purchase_date))                 7  \n",
       "NUM_UNIQUE(temp_new_trans.HOUR(purchase_date))                   13  \n",
       "MODE(temp_new_trans.DAY(purchase_date))                          24  \n",
       "MODE(temp_new_trans.MONTH(purchase_date))                        12  \n",
       "MODE(temp_new_trans.WEEKDAY(purchase_date))                       5  \n",
       "MODE(temp_new_trans.HOUR(purchase_date))                         11  \n",
       "TREND(temp_new_trans.PERCENTILE(purchased_on_we...      6.36819e-08  \n",
       "TREND(temp_new_trans.PERCENTILE(duration), purc...      5.40778e-09  \n",
       "TREND(temp_new_trans.PERCENTILE(month_diff), pu...     -1.98171e-08  \n",
       "TREND(temp_new_trans.PERCENTILE(amount_month_ra...       2.4331e-08  \n",
       "TREND(temp_new_trans.PERCENTILE(installments), ...                0  \n",
       "TREND(temp_new_trans.DAYS_SINCE(purchase_date),...     -1.15842e-05  \n",
       "TREND(temp_new_trans.PERCENTILE(purchase_amount...      1.73688e-08  \n",
       "TREND(temp_new_trans.PERCENTILE(month_lag), pur...      1.63584e-07  \n",
       "TREND(temp_new_trans.PERCENTILE(purchase_date_s...      8.74523e-08  \n",
       "PERCENTILE(SUM(temp_hist_trans.installments))                  0.75  \n",
       "PERCENTILE(SUM(temp_hist_trans.month_lag))                        1  \n",
       "PERCENTILE(SUM(temp_hist_trans.purchase_amount))                  1  \n",
       "PERCENTILE(SUM(temp_hist_trans.purchased_on_wee...              0.5  \n",
       "PERCENTILE(SUM(temp_hist_trans.month_diff))                     0.5  \n",
       "PERCENTILE(SUM(temp_hist_trans.purchase_date_su...              0.5  \n",
       "PERCENTILE(SUM(temp_hist_trans.duration))                         1  \n",
       "PERCENTILE(SUM(temp_hist_trans.amount_month_rat...                1  \n",
       "PERCENTILE(STD(temp_hist_trans.installments))                     1  \n",
       "PERCENTILE(STD(temp_hist_trans.month_lag))                      0.5  \n",
       "PERCENTILE(STD(temp_hist_trans.purchase_amount))                  1  \n",
       "PERCENTILE(STD(temp_hist_trans.purchased_on_wee...                1  \n",
       "PERCENTILE(STD(temp_hist_trans.month_diff))                       1  \n",
       "PERCENTILE(STD(temp_hist_trans.purchase_date_su...                1  \n",
       "PERCENTILE(STD(temp_hist_trans.duration))                         1  \n",
       "PERCENTILE(STD(temp_hist_trans.amount_month_rat...                1  \n",
       "PERCENTILE(MAX(temp_hist_trans.installments))                  0.75  \n",
       "PERCENTILE(MAX(temp_hist_trans.month_lag))                     0.75  \n",
       "PERCENTILE(MAX(temp_hist_trans.purchase_amount))                  1  \n",
       "PERCENTILE(MAX(temp_hist_trans.purchased_on_wee...             0.75  \n",
       "PERCENTILE(MAX(temp_hist_trans.month_diff))                       1  \n",
       "PERCENTILE(MAX(temp_hist_trans.purchase_date_su...                1  \n",
       "PERCENTILE(MAX(temp_hist_trans.duration))                         1  \n",
       "PERCENTILE(MAX(temp_hist_trans.amount_month_rat...              0.5  \n",
       "PERCENTILE(SKEW(temp_hist_trans.installments))                  0.5  \n",
       "PERCENTILE(SKEW(temp_hist_trans.month_lag))                     0.5  \n",
       "PERCENTILE(SKEW(temp_hist_trans.purchase_amount))               0.5  \n",
       "PERCENTILE(SKEW(temp_hist_trans.purchased_on_we...                1  \n",
       "PERCENTILE(SKEW(temp_hist_trans.month_diff))                    0.5  \n",
       "PERCENTILE(SKEW(temp_hist_trans.purchase_date_s...                1  \n",
       "PERCENTILE(SKEW(temp_hist_trans.duration))                      0.5  \n",
       "PERCENTILE(SKEW(temp_hist_trans.amount_month_ra...              0.5  \n",
       "PERCENTILE(MIN(temp_hist_trans.installments))                  0.75  \n",
       "PERCENTILE(MIN(temp_hist_trans.month_lag))                        1  \n",
       "PERCENTILE(MIN(temp_hist_trans.purchase_amount))                0.5  \n",
       "PERCENTILE(MIN(temp_hist_trans.purchased_on_wee...             0.75  \n",
       "PERCENTILE(MIN(temp_hist_trans.month_diff))                       1  \n",
       "PERCENTILE(MIN(temp_hist_trans.purchase_date_su...             0.75  \n",
       "PERCENTILE(MIN(temp_hist_trans.duration))                       0.5  \n",
       "PERCENTILE(MIN(temp_hist_trans.amount_month_rat...                1  \n",
       "PERCENTILE(MEAN(temp_hist_trans.installments))                    1  \n",
       "PERCENTILE(MEAN(temp_hist_trans.month_lag))                       1  \n",
       "PERCENTILE(MEAN(temp_hist_trans.purchase_amount))                 1  \n",
       "PERCENTILE(MEAN(temp_hist_trans.purchased_on_we...              0.5  \n",
       "PERCENTILE(MEAN(temp_hist_trans.month_diff))                      1  \n",
       "PERCENTILE(MEAN(temp_hist_trans.purchase_date_s...                1  \n",
       "PERCENTILE(MEAN(temp_hist_trans.duration))                      0.5  \n",
       "PERCENTILE(MEAN(temp_hist_trans.amount_month_ra...                1  \n",
       "PERCENTILE(COUNT(temp_hist_trans))                              0.5  \n",
       "PERCENTILE(MEDIAN(temp_hist_trans.installments))               0.75  \n",
       "PERCENTILE(MEDIAN(temp_hist_trans.month_lag))                     1  \n",
       "PERCENTILE(MEDIAN(temp_hist_trans.purchase_amou...              0.5  \n",
       "PERCENTILE(MEDIAN(temp_hist_trans.purchased_on_...             0.75  \n",
       "PERCENTILE(MEDIAN(temp_hist_trans.month_diff))                    1  \n",
       "PERCENTILE(MEDIAN(temp_hist_trans.purchase_date...              0.5  \n",
       "PERCENTILE(MEDIAN(temp_hist_trans.duration))                    0.5  \n",
       "PERCENTILE(MEDIAN(temp_hist_trans.amount_month_...                1  \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.authorize...             0.75  \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.city_id))                 0.5  \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.category_1))             0.75  \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.category_3))             0.75  \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.merchant_...              0.5  \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.merchant_...              0.5  \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.category_2))             0.75  \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.state_id))                0.5  \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.subsector...              0.5  \n",
       "PERCENTILE(NUM_UNIQUE(temp_hist_trans.purchased...             0.75  \n",
       "PERCENTILE(AVG_TIME_BETWEEN(temp_hist_trans.pur...                1  \n",
       "PERCENTILE(TIME_SINCE_LAST(temp_hist_trans.purc...                1  \n",
       "PERCENTILE(TREND(temp_hist_trans.month_diff, pu...              0.5  \n",
       "PERCENTILE(TREND(temp_hist_trans.installments, ...                1  \n",
       "PERCENTILE(TREND(temp_hist_trans.amount_month_r...              0.5  \n",
       "PERCENTILE(TREND(temp_hist_trans.purchase_amoun...              0.5  \n",
       "PERCENTILE(TREND(temp_hist_trans.purchase_date_...                1  \n",
       "PERCENTILE(TREND(temp_hist_trans.duration, purc...              0.5  \n",
       "PERCENTILE(TREND(temp_hist_trans.month_lag, pur...              0.5  \n",
       "PERCENTILE(TREND(temp_hist_trans.purchased_on_w...                1  \n",
       "PERCENTILE(SUM(temp_new_trans.installments))                   0.75  \n",
       "PERCENTILE(SUM(temp_new_trans.month_lag))                       0.5  \n",
       "PERCENTILE(SUM(temp_new_trans.purchase_amount))                   1  \n",
       "PERCENTILE(SUM(temp_new_trans.purchased_on_week...              0.5  \n",
       "PERCENTILE(SUM(temp_new_trans.month_diff))                        1  \n",
       "PERCENTILE(SUM(temp_new_trans.purchase_date_suc...              0.5  \n",
       "PERCENTILE(SUM(temp_new_trans.duration))                          1  \n",
       "PERCENTILE(SUM(temp_new_trans.amount_month_ratio))                1  \n",
       "PERCENTILE(STD(temp_new_trans.installments))                   0.75  \n",
       "PERCENTILE(STD(temp_new_trans.month_lag))                       0.5  \n",
       "PERCENTILE(STD(temp_new_trans.purchase_amount))                   1  \n",
       "PERCENTILE(STD(temp_new_trans.purchased_on_week...                1  \n",
       "PERCENTILE(STD(temp_new_trans.month_diff))                      0.5  \n",
       "PERCENTILE(STD(temp_new_trans.purchase_date_suc...                1  \n",
       "PERCENTILE(STD(temp_new_trans.duration))                          1  \n",
       "PERCENTILE(STD(temp_new_trans.amount_month_ratio))                1  \n",
       "PERCENTILE(MAX(temp_new_trans.installments))                   0.75  \n",
       "PERCENTILE(MAX(temp_new_trans.month_lag))                      0.75  \n",
       "PERCENTILE(MAX(temp_new_trans.purchase_amount))                   1  \n",
       "PERCENTILE(MAX(temp_new_trans.purchased_on_week...             0.75  \n",
       "PERCENTILE(MAX(temp_new_trans.month_diff))                        1  \n",
       "PERCENTILE(MAX(temp_new_trans.purchase_date_suc...              0.5  \n",
       "PERCENTILE(MAX(temp_new_trans.duration))                          1  \n",
       "PERCENTILE(MAX(temp_new_trans.amount_month_ratio))                1  \n",
       "PERCENTILE(SKEW(temp_new_trans.installments))                  0.75  \n",
       "PERCENTILE(SKEW(temp_new_trans.month_lag))                        1  \n",
       "PERCENTILE(SKEW(temp_new_trans.purchase_amount))                  1  \n",
       "PERCENTILE(SKEW(temp_new_trans.purchased_on_wee...                1  \n",
       "PERCENTILE(SKEW(temp_new_trans.month_diff))                       1  \n",
       "PERCENTILE(SKEW(temp_new_trans.purchase_date_su...              0.5  \n",
       "PERCENTILE(SKEW(temp_new_trans.duration))                         1  \n",
       "PERCENTILE(SKEW(temp_new_trans.amount_month_rat...                1  \n",
       "PERCENTILE(MIN(temp_new_trans.installments))                   0.75  \n",
       "PERCENTILE(MIN(temp_new_trans.month_lag))                      0.75  \n",
       "PERCENTILE(MIN(temp_new_trans.purchase_amount))                 0.5  \n",
       "PERCENTILE(MIN(temp_new_trans.purchased_on_week...             0.75  \n",
       "PERCENTILE(MIN(temp_new_trans.month_diff))                        1  \n",
       "PERCENTILE(MIN(temp_new_trans.purchase_date_suc...             0.75  \n",
       "PERCENTILE(MIN(temp_new_trans.duration))                        0.5  \n",
       "PERCENTILE(MIN(temp_new_trans.amount_month_ratio))                1  \n",
       "PERCENTILE(MEAN(temp_new_trans.installments))                  0.75  \n",
       "PERCENTILE(MEAN(temp_new_trans.month_lag))                      0.5  \n",
       "PERCENTILE(MEAN(temp_new_trans.purchase_amount))                  1  \n",
       "PERCENTILE(MEAN(temp_new_trans.purchased_on_wee...              0.5  \n",
       "PERCENTILE(MEAN(temp_new_trans.month_diff))                       1  \n",
       "PERCENTILE(MEAN(temp_new_trans.purchase_date_su...              0.5  \n",
       "PERCENTILE(MEAN(temp_new_trans.duration))                         1  \n",
       "PERCENTILE(MEAN(temp_new_trans.amount_month_rat...                1  \n",
       "PERCENTILE(COUNT(temp_new_trans))                                 1  \n",
       "PERCENTILE(MEDIAN(temp_new_trans.installments))                0.75  \n",
       "PERCENTILE(MEDIAN(temp_new_trans.month_lag))                   0.75  \n",
       "PERCENTILE(MEDIAN(temp_new_trans.purchase_amount))              0.5  \n",
       "PERCENTILE(MEDIAN(temp_new_trans.purchased_on_w...              0.5  \n",
       "PERCENTILE(MEDIAN(temp_new_trans.month_diff))                     1  \n",
       "PERCENTILE(MEDIAN(temp_new_trans.purchase_date_...              0.5  \n",
       "PERCENTILE(MEDIAN(temp_new_trans.duration))                     0.5  \n",
       "PERCENTILE(MEDIAN(temp_new_trans.amount_month_r...                1  \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.authorized...             0.75  \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.city_id))                    1  \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.category_1))              0.75  \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.category_3))              0.75  \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.merchant_c...             0.75  \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.merchant_id))                1  \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.category_2))                 1  \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.state_id))                   1  \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.subsector_...                1  \n",
       "PERCENTILE(NUM_UNIQUE(temp_new_trans.purchased_...             0.75  \n",
       "PERCENTILE(AVG_TIME_BETWEEN(temp_new_trans.purc...              0.5  \n",
       "PERCENTILE(TIME_SINCE_LAST(temp_new_trans.purch...                1  \n",
       "PERCENTILE(TREND(temp_new_trans.installments, p...             0.75  \n",
       "PERCENTILE(TREND(temp_new_trans.amount_month_ra...              0.5  \n",
       "PERCENTILE(TREND(temp_new_trans.purchase_amount...              0.5  \n",
       "PERCENTILE(TREND(temp_new_trans.purchase_date_s...                1  \n",
       "PERCENTILE(TREND(temp_new_trans.duration, purch...              0.5  \n",
       "PERCENTILE(TREND(temp_new_trans.purchased_on_we...                1  \n",
       "PERCENTILE(TREND(temp_new_trans.month_diff, pur...                1  \n",
       "PERCENTILE(TREND(temp_new_trans.month_lag, purc...                1  \n",
       "PERCENTILE(DAYS_SINCE(first_active_month))                     0.75  "
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "feature_matrix.T.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_matrix.target.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def additional_feats2(hist_df):\n",
    "    hist_df['price'] = hist_df['purchase_amount'] / hist_df['installments']\n",
    "    #Christmas : December 25 2017\n",
    "    hist_df['Christmas_Day_2017']=(pd.to_datetime('2017-12-25')-hist_df['purchase_date']).dt.days.apply(lambda x: x if x > 0 and x < 100 else 0)\n",
    "    #Mothers Day: May 14 2017\n",
    "    hist_df['Mothers_Day_2017']=(pd.to_datetime('2017-06-04')-hist_df['purchase_date']).dt.days.apply(lambda x: x if x > 0 and x < 100 else 0)\n",
    "    #fathers day: August 13 2017\n",
    "    hist_df['fathers_day_2017']=(pd.to_datetime('2017-08-13')-hist_df['purchase_date']).dt.days.apply(lambda x: x if x > 0 and x < 100 else 0)\n",
    "    #Childrens day: October 12 2017\n",
    "    hist_df['Children_day_2017']=(pd.to_datetime('2017-10-12')-hist_df['purchase_date']).dt.days.apply(lambda x: x if x > 0 and x < 100 else 0)\n",
    "    #Valentine's Day : 12th June, 2017\n",
    "    hist_df['Valentine_Day_2017']=(pd.to_datetime('2017-06-12')-hist_df['purchase_date']).dt.days.apply(lambda x: x if x > 0 and x < 100 else 0)\n",
    "    #Black Friday : 24th November 2017\n",
    "    hist_df['Black_Friday_2017']=(pd.to_datetime('2017-11-24') - hist_df['purchase_date']).dt.days.apply(lambda x: x if x > 0 and x < 100 else 0)\n",
    "    #2018\n",
    "    #Mothers Day: May 13 2018\n",
    "    hist_df['Mothers_Day_2018']=(pd.to_datetime('2018-05-13')-hist_df['purchase_date']).dt.days.apply(lambda x: x if x > 0 and x < 100 else 0)\n",
    "    return hist_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "temp_hist_trans.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "hist_trans2, new_hist_trans2 = [additional_feats2(df) for df in [temp_hist_trans, temp_new_trans]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "dfs2 = [hist_trans2, new_hist_trans2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def aggregate_by_card_id(df):\n",
    "    aggs = {}\n",
    "    aggs['Christmas_Day_2017'] = ['mean']\n",
    "    aggs['Mothers_Day_2017'] = ['mean']\n",
    "    aggs['fathers_day_2017'] = ['mean']\n",
    "    aggs['Children_day_2017'] = ['mean']\n",
    "    aggs['Valentine_Day_2017'] = ['mean']\n",
    "    aggs['Black_Friday_2017'] = ['mean']\n",
    "    aggs['Mothers_Day_2018'] = ['mean']\n",
    "    aggs['price'] = ['sum','mean','max','min','var', 'skew']\n",
    "    aggs['purchase_date'] = ['max','min']\n",
    "    aggs['card_id'] = ['size']\n",
    "    aggs['category_1'] = ['sum', 'mean']\n",
    "    aggs['category_2'] = ['mean']\n",
    "    aggs['category_3'] = ['mean']\n",
    "    new_df = df.groupby(['card_id']).agg(aggs)\n",
    "    new_df.columns = ['_'.join(col).strip() for col in new_df.columns.values]\n",
    "    new_df.reset_index(inplace=True)\n",
    "    for col in ['category_2','category_3', 'subsector_id', 'state_id']:\n",
    "        df[col+'_mean'] = df.groupby([col])['purchase_amount'].transform('mean')\n",
    "        df[col+'_min'] = df.groupby([col])['purchase_amount'].transform('min')\n",
    "        df[col+'_max'] = df.groupby([col])['purchase_amount'].transform('max')\n",
    "        df[col+'_sum'] = df.groupby([col])['purchase_amount'].transform('sum')\n",
    "        aggs[col+'_mean'] = ['mean']\n",
    "\n",
    "    other_df = (df.groupby('card_id')\n",
    "          .size()\n",
    "          .reset_index(name='transactions_count'))\n",
    "    \n",
    "    new_df = pd.merge(other_df, new_df, on='card_id', how='left')\n",
    "\n",
    "    new_df['purchase_date_diff'] = (new_df['purchase_date_max'] - new_df['purchase_date_min']).dt.days\n",
    "    new_df['purchase_date_average'] = new_df['purchase_date_diff']/new_df['card_id_size']\n",
    "    new_df['purchase_date_uptonow'] = (datetime.datetime.today() - new_df['purchase_date_max']).dt.days\n",
    "    new_df['purchase_date_uptomin'] = (datetime.datetime.today() - new_df['purchase_date_min']).dt.days\n",
    "    new_df['inverse_avg_transactions_per_day'] = new_df['purchase_date_diff']/new_df['card_id_size']\n",
    "    new_df['days_since_last_transaction'] = (datetime.datetime.today() - new_df['purchase_date_max']).dt.days\n",
    "    return new_df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "gc.collect()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "%time hist_trans_agg, new_hist_trans_agg = [aggregate_by_card_id(df) for df in dfs2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "hist_trans_agg.shape, new_hist_trans_agg.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "feature_matrix.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "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>card_id</th>\n",
       "      <th>transactions_count</th>\n",
       "      <th>Christmas_Day_2017_mean</th>\n",
       "      <th>Mothers_Day_2017_mean</th>\n",
       "      <th>fathers_day_2017_mean</th>\n",
       "      <th>Children_day_2017_mean</th>\n",
       "      <th>Valentine_Day_2017_mean</th>\n",
       "      <th>Black_Friday_2017_mean</th>\n",
       "      <th>Mothers_Day_2018_mean</th>\n",
       "      <th>price_sum</th>\n",
       "      <th>price_mean</th>\n",
       "      <th>price_max</th>\n",
       "      <th>price_min</th>\n",
       "      <th>price_var</th>\n",
       "      <th>price_skew</th>\n",
       "      <th>purchase_date_max</th>\n",
       "      <th>purchase_date_min</th>\n",
       "      <th>card_id_size</th>\n",
       "      <th>category_1_sum</th>\n",
       "      <th>category_1_mean</th>\n",
       "      <th>category_2_mean</th>\n",
       "      <th>category_3_mean</th>\n",
       "      <th>purchase_date_diff</th>\n",
       "      <th>purchase_date_average</th>\n",
       "      <th>purchase_date_uptonow</th>\n",
       "      <th>purchase_date_uptomin</th>\n",
       "      <th>inverse_avg_transactions_per_day</th>\n",
       "      <th>days_since_last_transaction</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>C_ID_2223b33279</td>\n",
       "      <td>21</td>\n",
       "      <td>12.761905</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>-inf</td>\n",
       "      <td>-inf</td>\n",
       "      <td>-inf</td>\n",
       "      <td>-inf</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2018-01-25 12:23:25</td>\n",
       "      <td>2017-12-01 12:03:54</td>\n",
       "      <td>21</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.809524</td>\n",
       "      <td>0</td>\n",
       "      <td>55</td>\n",
       "      <td>2.619048</td>\n",
       "      <td>377</td>\n",
       "      <td>432</td>\n",
       "      <td>2.619048</td>\n",
       "      <td>377</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>C_ID_92a2005557</td>\n",
       "      <td>23</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>41.73913</td>\n",
       "      <td>-inf</td>\n",
       "      <td>-inf</td>\n",
       "      <td>-inf</td>\n",
       "      <td>-inf</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2018-04-29 11:23:05</td>\n",
       "      <td>2018-03-05 14:04:36</td>\n",
       "      <td>23</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0</td>\n",
       "      <td>54</td>\n",
       "      <td>2.347826</td>\n",
       "      <td>283</td>\n",
       "      <td>338</td>\n",
       "      <td>2.347826</td>\n",
       "      <td>283</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>C_ID_ae86f7d5fb</td>\n",
       "      <td>24</td>\n",
       "      <td>4.250000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>inf</td>\n",
       "      <td>-inf</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2018-01-26 10:02:57</td>\n",
       "      <td>2017-12-02 16:32:02</td>\n",
       "      <td>24</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2.833333</td>\n",
       "      <td>0</td>\n",
       "      <td>54</td>\n",
       "      <td>2.250000</td>\n",
       "      <td>376</td>\n",
       "      <td>431</td>\n",
       "      <td>2.250000</td>\n",
       "      <td>376</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           card_id  transactions_count  Christmas_Day_2017_mean  \\\n",
       "0  C_ID_2223b33279                  21                12.761905   \n",
       "1  C_ID_92a2005557                  23                 0.000000   \n",
       "2  C_ID_ae86f7d5fb                  24                 4.250000   \n",
       "\n",
       "   Mothers_Day_2017_mean  fathers_day_2017_mean  Children_day_2017_mean  \\\n",
       "0                      0                      0                       0   \n",
       "1                      0                      0                       0   \n",
       "2                      0                      0                       0   \n",
       "\n",
       "   Valentine_Day_2017_mean  Black_Friday_2017_mean  Mothers_Day_2018_mean  \\\n",
       "0                        0                       0                0.00000   \n",
       "1                        0                       0               41.73913   \n",
       "2                        0                       0                0.00000   \n",
       "\n",
       "   price_sum  price_mean  price_max  price_min  price_var  price_skew  \\\n",
       "0       -inf        -inf       -inf       -inf        NaN         NaN   \n",
       "1       -inf        -inf       -inf       -inf        NaN         NaN   \n",
       "2        NaN         NaN        inf       -inf        NaN         NaN   \n",
       "\n",
       "    purchase_date_max   purchase_date_min  card_id_size  category_1_sum  \\\n",
       "0 2018-01-25 12:23:25 2017-12-01 12:03:54            21               0   \n",
       "1 2018-04-29 11:23:05 2018-03-05 14:04:36            23               0   \n",
       "2 2018-01-26 10:02:57 2017-12-02 16:32:02            24               0   \n",
       "\n",
       "   category_1_mean  category_2_mean  category_3_mean  purchase_date_diff  \\\n",
       "0                0         1.809524                0                  55   \n",
       "1                0         1.000000                0                  54   \n",
       "2                0         2.833333                0                  54   \n",
       "\n",
       "   purchase_date_average  purchase_date_uptonow  purchase_date_uptomin  \\\n",
       "0               2.619048                    377                    432   \n",
       "1               2.347826                    283                    338   \n",
       "2               2.250000                    376                    431   \n",
       "\n",
       "   inverse_avg_transactions_per_day  days_since_last_transaction  \n",
       "0                          2.619048                          377  \n",
       "1                          2.347826                          283  \n",
       "2                          2.250000                          376  "
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "new_hist_trans_agg.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def add_extra_cols_on_agg(df):\n",
    "    df['repurchase_merchant_rate_hist'] = df['COUNT(temp_hist_trans)']/df['NUM_UNIQUE(temp_hist_trans.merchant_id)']\n",
    "    df['repurchase_merchant_rate_new'] = df['COUNT(temp_new_trans)']/df['NUM_UNIQUE(temp_new_trans.merchant_id)']\n",
    "\n",
    "    df['merchant_category_repurchase_new'] = df['NUM_UNIQUE(temp_new_trans.merchant_category_id)']/df['NUM_UNIQUE(temp_new_trans.merchant_id)']\n",
    "    df['merchant_category_repurchase_hist'] = df['NUM_UNIQUE(temp_hist_trans.merchant_category_id)']/df['NUM_UNIQUE(temp_hist_trans.merchant_id)']\n",
    "\n",
    "    df['avg_spend_per_merchant_hist'] = df['SUM(temp_hist_trans.purchase_amount)']/df['NUM_UNIQUE(temp_hist_trans.merchant_id)']\n",
    "    df['avg_spend_per_merchant_new'] = df['SUM(temp_new_trans.purchase_amount)']/df['NUM_UNIQUE(temp_new_trans.merchant_id)']\n",
    "\n",
    "    df['avg_trans_per_merchant_new'] = df['COUNT(temp_new_trans)']/df['NUM_UNIQUE(temp_new_trans.merchant_id)']\n",
    "    df['avg_trans_per_merchant_hist'] = df['COUNT(temp_hist_trans)']/df['NUM_UNIQUE(temp_hist_trans.merchant_id)']\n",
    "\n",
    "    df['avg_spend_per_transaction_hist'] = df['SUM(temp_hist_trans.purchase_amount)']/df['COUNT(temp_hist_trans)']\n",
    "    df['avg_spend_per_transaction_new'] = df['SUM(temp_new_trans.purchase_amount)']/df['COUNT(temp_new_trans)']\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "scrolled": false
   },
   "outputs": [],
   "source": [
    "df_fm = add_extra_cols_on_agg(feature_matrix)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_fm.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "heading_collapsed": true
   },
   "source": [
    "### Interactions in test&train dfs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "l_es = ft.EntitySet()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    " test['feature_3'] = test['feature_3'].map({0:1, 1:2})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "hidden": true
   },
   "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>first_active_month</th>\n",
       "      <th>card_id</th>\n",
       "      <th>feature_1</th>\n",
       "      <th>feature_2</th>\n",
       "      <th>feature_3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2017-04</td>\n",
       "      <td>C_ID_0ab67a22ab</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017-01</td>\n",
       "      <td>C_ID_130fd0cbdd</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2017-08</td>\n",
       "      <td>C_ID_b709037bc5</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2017-12</td>\n",
       "      <td>C_ID_d27d835a9f</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2015-12</td>\n",
       "      <td>C_ID_2b5e3df5c2</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  first_active_month          card_id  feature_1  feature_2  feature_3\n",
       "0            2017-04  C_ID_0ab67a22ab          3          3          2\n",
       "1            2017-01  C_ID_130fd0cbdd          2          3          1\n",
       "2            2017-08  C_ID_b709037bc5          5          1          2\n",
       "3            2017-12  C_ID_d27d835a9f          2          1          1\n",
       "4            2015-12  C_ID_2b5e3df5c2          5          1          2"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# lala_cols = {}\n",
    "# lala_cols['feature_3'] = ft.variable_types.variable.Boolean"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'feature_3': featuretools.variable_types.variable.Boolean}"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# lala_cols"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'temp_train' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-27-badd42e630bd>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mlala_es\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0ml_es\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mentity_from_dataframe\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'train'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtemp_train\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;34m'card_id'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mtime_index\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'first_active_month'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m: name 'temp_train' is not defined"
     ]
    }
   ],
   "source": [
    "lala_es = l_es.entity_from_dataframe('train', temp_train, 'card_id', time_index='first_active_month')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 101,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(123623, 5)"
      ]
     },
     "execution_count": 101,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "test.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "lala_es = l_es.entity_from_dataframe('test', test, 'card_id', time_index='first_active_month')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Entity: test\n",
       "  Variables:\n",
       "    card_id (dtype: index)\n",
       "    first_active_month (dtype: datetime_time_index)\n",
       "    feature_1 (dtype: numeric)\n",
       "    feature_2 (dtype: numeric)\n",
       "    feature_3 (dtype: numeric)\n",
       "  Shape:\n",
       "    (Rows: 123623, Columns: 5)"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lala_es['test']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "# lala_es['test']['feature_3'].interesting_values = [1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Built 1426 features\n",
      "Elapsed: 00:18 | Remaining: 00:00 | Progress: 100%|██████████| Calculated: 11/11 chunks\n"
     ]
    }
   ],
   "source": [
    "lala_feature_matrix, lala_feature_names = ft.dfs(entityset = lala_es, target_entity = 'test',\n",
    "                                       trans_primitives = default_trans_primitives,\n",
    "                                       agg_primitives=default_agg_primitives,\n",
    "                                       where_primitives = ['mean', 'sum'],\n",
    "                                        max_depth = 3, features_only=False, verbose = True)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "hidden": true
   },
   "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>first_active_month</th>\n",
       "      <th>card_id</th>\n",
       "      <th>feature_1</th>\n",
       "      <th>feature_2</th>\n",
       "      <th>feature_3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>123623</td>\n",
       "      <td>123623</td>\n",
       "      <td>123623</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3.10926</td>\n",
       "      <td>1.7418</td>\n",
       "      <td>1.56438</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.18911</td>\n",
       "      <td>0.749195</td>\n",
       "      <td>0.49584</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>counts</th>\n",
       "      <td>123622</td>\n",
       "      <td>123623</td>\n",
       "      <td>123623</td>\n",
       "      <td>123623</td>\n",
       "      <td>123623</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>uniques</th>\n",
       "      <td>75</td>\n",
       "      <td>123623</td>\n",
       "      <td>5</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>missing</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>missing_perc</th>\n",
       "      <td>0.00%</td>\n",
       "      <td>0%</td>\n",
       "      <td>0%</td>\n",
       "      <td>0%</td>\n",
       "      <td>0%</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>types</th>\n",
       "      <td>date</td>\n",
       "      <td>unique</td>\n",
       "      <td>numeric</td>\n",
       "      <td>numeric</td>\n",
       "      <td>bool</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             first_active_month card_id feature_1 feature_2 feature_3\n",
       "count                       NaN     NaN    123623    123623    123623\n",
       "mean                        NaN     NaN   3.10926    1.7418   1.56438\n",
       "std                         NaN     NaN   1.18911  0.749195   0.49584\n",
       "min                         NaN     NaN         1         1         1\n",
       "25%                         NaN     NaN         2         1         1\n",
       "50%                         NaN     NaN         3         2         2\n",
       "75%                         NaN     NaN         4         2         2\n",
       "max                         NaN     NaN         5         3         2\n",
       "counts                   123622  123623    123623    123623    123623\n",
       "uniques                      75  123623         5         3         2\n",
       "missing                       1       0         0         0         0\n",
       "missing_perc              0.00%      0%        0%        0%        0%\n",
       "types                      date  unique   numeric   numeric      bool"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "DataFrameSummary(test).summary()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "hidden": true
   },
   "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_1</th>\n",
       "      <th>feature_2</th>\n",
       "      <th>feature_3</th>\n",
       "      <th>DAYS_SINCE(first_active_month)</th>\n",
       "      <th>feature_1 + feature_3</th>\n",
       "      <th>feature_1 + feature_2</th>\n",
       "      <th>feature_2 + feature_3</th>\n",
       "      <th>feature_3 % feature_1</th>\n",
       "      <th>feature_3 % feature_2</th>\n",
       "      <th>feature_1 % feature_3</th>\n",
       "      <th>...</th>\n",
       "      <th>feature_1 % feature_2 + feature_3 * feature_2 + feature_3 % feature_1 + feature_3</th>\n",
       "      <th>feature_2 + feature_3 % feature_1 + feature_3 * feature_3 % feature_1 + feature_3</th>\n",
       "      <th>DAYS_SINCE(first_active_month) % feature_1 * feature_3 % DAYS_SINCE(first_active_month)</th>\n",
       "      <th>DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_2 + feature_3 % feature_1 + feature_3</th>\n",
       "      <th>feature_1 % DAYS_SINCE(first_active_month) * feature_1 % feature_2</th>\n",
       "      <th>feature_2 + feature_3 % feature_1 + feature_3 * feature_3</th>\n",
       "      <th>feature_2 % feature_2 + feature_3 * feature_3 % feature_2 + feature_3</th>\n",
       "      <th>feature_2 % feature_1 + feature_2 * feature_3 % DAYS_SINCE(first_active_month)</th>\n",
       "      <th>feature_1 % feature_1 + feature_3 * feature_2 + feature_3 % feature_1 + feature_3</th>\n",
       "      <th>DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_3 % feature_1 + feature_3</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>card_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>C_ID_0001238066</th>\n",
       "      <td>4.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>522.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C_ID_0001793786</th>\n",
       "      <td>5.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>765.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>4.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C_ID_00024e244b</th>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>918.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C_ID_0002709b5a</th>\n",
       "      <td>5.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1284.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C_ID_000298032a</th>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>796.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 1426 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 feature_1  feature_2  feature_3  \\\n",
       "card_id                                            \n",
       "C_ID_0001238066        4.0        1.0        1.0   \n",
       "C_ID_0001793786        5.0        2.0        2.0   \n",
       "C_ID_00024e244b        2.0        2.0        1.0   \n",
       "C_ID_0002709b5a        5.0        1.0        2.0   \n",
       "C_ID_000298032a        2.0        2.0        1.0   \n",
       "\n",
       "                 DAYS_SINCE(first_active_month)  feature_1 + feature_3  \\\n",
       "card_id                                                                  \n",
       "C_ID_0001238066                           522.0                    5.0   \n",
       "C_ID_0001793786                           765.0                    7.0   \n",
       "C_ID_00024e244b                           918.0                    3.0   \n",
       "C_ID_0002709b5a                          1284.0                    7.0   \n",
       "C_ID_000298032a                           796.0                    3.0   \n",
       "\n",
       "                 feature_1 + feature_2  feature_2 + feature_3  \\\n",
       "card_id                                                         \n",
       "C_ID_0001238066                    5.0                    2.0   \n",
       "C_ID_0001793786                    7.0                    4.0   \n",
       "C_ID_00024e244b                    4.0                    3.0   \n",
       "C_ID_0002709b5a                    6.0                    3.0   \n",
       "C_ID_000298032a                    4.0                    3.0   \n",
       "\n",
       "                 feature_3 % feature_1  feature_3 % feature_2  \\\n",
       "card_id                                                         \n",
       "C_ID_0001238066                    1.0                    0.0   \n",
       "C_ID_0001793786                    2.0                    0.0   \n",
       "C_ID_00024e244b                    1.0                    1.0   \n",
       "C_ID_0002709b5a                    2.0                    0.0   \n",
       "C_ID_000298032a                    1.0                    1.0   \n",
       "\n",
       "                 feature_1 % feature_3  \\\n",
       "card_id                                  \n",
       "C_ID_0001238066                    0.0   \n",
       "C_ID_0001793786                    1.0   \n",
       "C_ID_00024e244b                    0.0   \n",
       "C_ID_0002709b5a                    1.0   \n",
       "C_ID_000298032a                    0.0   \n",
       "\n",
       "                                                            ...                                              \\\n",
       "card_id                                                     ...                                               \n",
       "C_ID_0001238066                                             ...                                               \n",
       "C_ID_0001793786                                             ...                                               \n",
       "C_ID_00024e244b                                             ...                                               \n",
       "C_ID_0002709b5a                                             ...                                               \n",
       "C_ID_000298032a                                             ...                                               \n",
       "\n",
       "                 feature_1 % feature_2 + feature_3 * feature_2 + feature_3 % feature_1 + feature_3  \\\n",
       "card_id                                                                                              \n",
       "C_ID_0001238066                                                0.0                                   \n",
       "C_ID_0001793786                                                4.0                                   \n",
       "C_ID_00024e244b                                                0.0                                   \n",
       "C_ID_0002709b5a                                                6.0                                   \n",
       "C_ID_000298032a                                                0.0                                   \n",
       "\n",
       "                 feature_2 + feature_3 % feature_1 + feature_3 * feature_3 % feature_1 + feature_3  \\\n",
       "card_id                                                                                              \n",
       "C_ID_0001238066                                                2.0                                   \n",
       "C_ID_0001793786                                                8.0                                   \n",
       "C_ID_00024e244b                                                0.0                                   \n",
       "C_ID_0002709b5a                                                6.0                                   \n",
       "C_ID_000298032a                                                0.0                                   \n",
       "\n",
       "                 DAYS_SINCE(first_active_month) % feature_1 * feature_3 % DAYS_SINCE(first_active_month)  \\\n",
       "card_id                                                                                                    \n",
       "C_ID_0001238066                                                2.0                                         \n",
       "C_ID_0001793786                                                0.0                                         \n",
       "C_ID_00024e244b                                                0.0                                         \n",
       "C_ID_0002709b5a                                                8.0                                         \n",
       "C_ID_000298032a                                                0.0                                         \n",
       "\n",
       "                 DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_2 + feature_3 % feature_1 + feature_3  \\\n",
       "card_id                                                                                                                   \n",
       "C_ID_0001238066                                                4.0                                                        \n",
       "C_ID_0001793786                                                8.0                                                        \n",
       "C_ID_00024e244b                                                0.0                                                        \n",
       "C_ID_0002709b5a                                                9.0                                                        \n",
       "C_ID_000298032a                                                0.0                                                        \n",
       "\n",
       "                 feature_1 % DAYS_SINCE(first_active_month) * feature_1 % feature_2  \\\n",
       "card_id                                                                               \n",
       "C_ID_0001238066                                                0.0                    \n",
       "C_ID_0001793786                                                5.0                    \n",
       "C_ID_00024e244b                                                0.0                    \n",
       "C_ID_0002709b5a                                                0.0                    \n",
       "C_ID_000298032a                                                0.0                    \n",
       "\n",
       "                 feature_2 + feature_3 % feature_1 + feature_3 * feature_3  \\\n",
       "card_id                                                                      \n",
       "C_ID_0001238066                                                2.0           \n",
       "C_ID_0001793786                                                8.0           \n",
       "C_ID_00024e244b                                                0.0           \n",
       "C_ID_0002709b5a                                                6.0           \n",
       "C_ID_000298032a                                                0.0           \n",
       "\n",
       "                 feature_2 % feature_2 + feature_3 * feature_3 % feature_2 + feature_3  \\\n",
       "card_id                                                                                  \n",
       "C_ID_0001238066                                                1.0                       \n",
       "C_ID_0001793786                                                4.0                       \n",
       "C_ID_00024e244b                                                2.0                       \n",
       "C_ID_0002709b5a                                                2.0                       \n",
       "C_ID_000298032a                                                2.0                       \n",
       "\n",
       "                 feature_2 % feature_1 + feature_2 * feature_3 % DAYS_SINCE(first_active_month)  \\\n",
       "card_id                                                                                           \n",
       "C_ID_0001238066                                                1.0                                \n",
       "C_ID_0001793786                                                4.0                                \n",
       "C_ID_00024e244b                                                2.0                                \n",
       "C_ID_0002709b5a                                                2.0                                \n",
       "C_ID_000298032a                                                2.0                                \n",
       "\n",
       "                 feature_1 % feature_1 + feature_3 * feature_2 + feature_3 % feature_1 + feature_3  \\\n",
       "card_id                                                                                              \n",
       "C_ID_0001238066                                                8.0                                   \n",
       "C_ID_0001793786                                               20.0                                   \n",
       "C_ID_00024e244b                                                0.0                                   \n",
       "C_ID_0002709b5a                                               15.0                                   \n",
       "C_ID_000298032a                                                0.0                                   \n",
       "\n",
       "                 DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_3 % feature_1 + feature_3  \n",
       "card_id                                                                                                      \n",
       "C_ID_0001238066                                                2.0                                           \n",
       "C_ID_0001793786                                                4.0                                           \n",
       "C_ID_00024e244b                                                2.0                                           \n",
       "C_ID_0002709b5a                                                0.0                                           \n",
       "C_ID_000298032a                                                0.0                                           \n",
       "\n",
       "[5 rows x 1426 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lala_feature_matrix.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "default_agg_primitives =  [\"sum\", \"std\", \"max\", \"skew\", \"min\", \"mean\", \"count\", \"median\",\n",
    "                           \"percent_true\", \"num_unique\", \"mode\", \"avg_time_between\", \"time_since_last\",\n",
    "                           \"percent_true\", \"trend\"]\n",
    "default_trans_primitives =  [\"days_since\", \"add\", \"mod\", \"multiply\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {
    "collapsed": true,
    "hidden": true
   },
   "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>name</th>\n",
       "      <th>type</th>\n",
       "      <th>description</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>all</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Test if all values are 'True'.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>any</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Test if any value is 'True'.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>mode</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Finds the most common element in a categorical...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>last</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Returns the last value.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>mean</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Computes the average value of a numeric feature.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>std</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Finds the standard deviation of a numeric feat...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>num_true</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Finds the number of 'True' values in a boolean.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>sum</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Sums elements of a numeric or boolean feature.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>skew</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Computes the skewness of a data set.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>avg_time_between</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Computes the average time between consecutive ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>n_most_common</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Finds the N most common elements in a categori...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>num_unique</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Returns the number of unique categorical varia...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>count</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Counts the number of non null values.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>median</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Finds the median value of any feature with wel...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>max</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Finds the maximum non-null value of a numeric ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>percent_true</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Finds the percent of 'True' values in a boolea...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>time_since_last</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Time since last related instance.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>trend</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Calculates the slope of the linear trend of va...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>min</td>\n",
       "      <td>aggregation</td>\n",
       "      <td>Finds the minimum non-null value of a numeric ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>cum_count</td>\n",
       "      <td>transform</td>\n",
       "      <td>Calculates the number of previous values of an...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>multiply</td>\n",
       "      <td>transform</td>\n",
       "      <td>Creates a transform feature that multplies two...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>characters</td>\n",
       "      <td>transform</td>\n",
       "      <td>Return the characters in a given string.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>cum_sum</td>\n",
       "      <td>transform</td>\n",
       "      <td>Calculates the sum of previous values of an in...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>add</td>\n",
       "      <td>transform</td>\n",
       "      <td>Creates a transform feature that adds two feat...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>time_since_previous</td>\n",
       "      <td>transform</td>\n",
       "      <td>Compute the time since the previous instance.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>months</td>\n",
       "      <td>transform</td>\n",
       "      <td>Transform a Timedelta feature into the number ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>weekend</td>\n",
       "      <td>transform</td>\n",
       "      <td>Transform Datetime feature into the boolean of...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>and</td>\n",
       "      <td>transform</td>\n",
       "      <td>For two boolean values, determine if both valu...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>longitude</td>\n",
       "      <td>transform</td>\n",
       "      <td>Returns the second value on the tuple base fea...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>weeks</td>\n",
       "      <td>transform</td>\n",
       "      <td>Transform a Timedelta feature into the number ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>minutes</td>\n",
       "      <td>transform</td>\n",
       "      <td>Transform a Timedelta feature into the number ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>percentile</td>\n",
       "      <td>transform</td>\n",
       "      <td>For each value of the base feature, determines...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>diff</td>\n",
       "      <td>transform</td>\n",
       "      <td>Compute the difference between the value of a ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>seconds</td>\n",
       "      <td>transform</td>\n",
       "      <td>Transform a Timedelta feature into the number ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>negate</td>\n",
       "      <td>transform</td>\n",
       "      <td>Creates a transform feature that negates a fea...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>hours</td>\n",
       "      <td>transform</td>\n",
       "      <td>Transform a Timedelta feature into the number ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>days_since</td>\n",
       "      <td>transform</td>\n",
       "      <td>For each value of the base feature, compute th...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>cum_max</td>\n",
       "      <td>transform</td>\n",
       "      <td>Calculates the max of previous values of an in...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>divide</td>\n",
       "      <td>transform</td>\n",
       "      <td>Creates a transform feature that divides two f...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>day</td>\n",
       "      <td>transform</td>\n",
       "      <td>Transform a Datetime feature into the day.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>40</th>\n",
       "      <td>days</td>\n",
       "      <td>transform</td>\n",
       "      <td>Transform a Timedelta feature into the number ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>is_null</td>\n",
       "      <td>transform</td>\n",
       "      <td>For each value of base feature, return 'True' ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>42</th>\n",
       "      <td>numwords</td>\n",
       "      <td>transform</td>\n",
       "      <td>Returns the words in a given string by countin...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>43</th>\n",
       "      <td>cum_mean</td>\n",
       "      <td>transform</td>\n",
       "      <td>Calculates the mean of previous values of an i...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>44</th>\n",
       "      <td>subtract</td>\n",
       "      <td>transform</td>\n",
       "      <td>Creates a transform feature that subtracts two...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>year</td>\n",
       "      <td>transform</td>\n",
       "      <td>Transform a Datetime feature into the year.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>weekday</td>\n",
       "      <td>transform</td>\n",
       "      <td>Transform Datetime feature into the boolean of...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>47</th>\n",
       "      <td>or</td>\n",
       "      <td>transform</td>\n",
       "      <td>For two boolean values, determine if one value...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>48</th>\n",
       "      <td>absolute</td>\n",
       "      <td>transform</td>\n",
       "      <td>Absolute value of base feature.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49</th>\n",
       "      <td>haversine</td>\n",
       "      <td>transform</td>\n",
       "      <td>Calculate the approximate haversine distance i...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50</th>\n",
       "      <td>month</td>\n",
       "      <td>transform</td>\n",
       "      <td>Transform a Datetime feature into the month.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>51</th>\n",
       "      <td>latitude</td>\n",
       "      <td>transform</td>\n",
       "      <td>Returns the first value of the tuple base feat...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>week</td>\n",
       "      <td>transform</td>\n",
       "      <td>Transform a Datetime feature into the week.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>53</th>\n",
       "      <td>minute</td>\n",
       "      <td>transform</td>\n",
       "      <td>Transform a Datetime feature into the minute.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>54</th>\n",
       "      <td>not</td>\n",
       "      <td>transform</td>\n",
       "      <td>For each value of the base feature, negates th...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>55</th>\n",
       "      <td>time_since</td>\n",
       "      <td>transform</td>\n",
       "      <td>Calculates time since the cutoff time.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>56</th>\n",
       "      <td>isin</td>\n",
       "      <td>transform</td>\n",
       "      <td>For each value of the base feature, checks whe...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>57</th>\n",
       "      <td>second</td>\n",
       "      <td>transform</td>\n",
       "      <td>Transform a Datetime feature into the second.</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>58</th>\n",
       "      <td>cum_min</td>\n",
       "      <td>transform</td>\n",
       "      <td>Calculates the min of previous values of an in...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>59</th>\n",
       "      <td>mod</td>\n",
       "      <td>transform</td>\n",
       "      <td>Creates a transform feature that divides two f...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>60</th>\n",
       "      <td>years</td>\n",
       "      <td>transform</td>\n",
       "      <td>Transform a Timedelta feature into the number ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>61</th>\n",
       "      <td>hour</td>\n",
       "      <td>transform</td>\n",
       "      <td>Transform a Datetime feature into the hour.</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                   name         type  \\\n",
       "0                   all  aggregation   \n",
       "1                   any  aggregation   \n",
       "2                  mode  aggregation   \n",
       "3                  last  aggregation   \n",
       "4                  mean  aggregation   \n",
       "5                   std  aggregation   \n",
       "6              num_true  aggregation   \n",
       "7                   sum  aggregation   \n",
       "8                  skew  aggregation   \n",
       "9      avg_time_between  aggregation   \n",
       "10        n_most_common  aggregation   \n",
       "11           num_unique  aggregation   \n",
       "12                count  aggregation   \n",
       "13               median  aggregation   \n",
       "14                  max  aggregation   \n",
       "15         percent_true  aggregation   \n",
       "16      time_since_last  aggregation   \n",
       "17                trend  aggregation   \n",
       "18                  min  aggregation   \n",
       "19            cum_count    transform   \n",
       "20             multiply    transform   \n",
       "21           characters    transform   \n",
       "22              cum_sum    transform   \n",
       "23                  add    transform   \n",
       "24  time_since_previous    transform   \n",
       "25               months    transform   \n",
       "26              weekend    transform   \n",
       "27                  and    transform   \n",
       "28            longitude    transform   \n",
       "29                weeks    transform   \n",
       "30              minutes    transform   \n",
       "31           percentile    transform   \n",
       "32                 diff    transform   \n",
       "33              seconds    transform   \n",
       "34               negate    transform   \n",
       "35                hours    transform   \n",
       "36           days_since    transform   \n",
       "37              cum_max    transform   \n",
       "38               divide    transform   \n",
       "39                  day    transform   \n",
       "40                 days    transform   \n",
       "41              is_null    transform   \n",
       "42             numwords    transform   \n",
       "43             cum_mean    transform   \n",
       "44             subtract    transform   \n",
       "45                 year    transform   \n",
       "46              weekday    transform   \n",
       "47                   or    transform   \n",
       "48             absolute    transform   \n",
       "49            haversine    transform   \n",
       "50                month    transform   \n",
       "51             latitude    transform   \n",
       "52                 week    transform   \n",
       "53               minute    transform   \n",
       "54                  not    transform   \n",
       "55           time_since    transform   \n",
       "56                 isin    transform   \n",
       "57               second    transform   \n",
       "58              cum_min    transform   \n",
       "59                  mod    transform   \n",
       "60                years    transform   \n",
       "61                 hour    transform   \n",
       "\n",
       "                                          description  \n",
       "0                      Test if all values are 'True'.  \n",
       "1                        Test if any value is 'True'.  \n",
       "2   Finds the most common element in a categorical...  \n",
       "3                             Returns the last value.  \n",
       "4    Computes the average value of a numeric feature.  \n",
       "5   Finds the standard deviation of a numeric feat...  \n",
       "6     Finds the number of 'True' values in a boolean.  \n",
       "7      Sums elements of a numeric or boolean feature.  \n",
       "8                Computes the skewness of a data set.  \n",
       "9   Computes the average time between consecutive ...  \n",
       "10  Finds the N most common elements in a categori...  \n",
       "11  Returns the number of unique categorical varia...  \n",
       "12              Counts the number of non null values.  \n",
       "13  Finds the median value of any feature with wel...  \n",
       "14  Finds the maximum non-null value of a numeric ...  \n",
       "15  Finds the percent of 'True' values in a boolea...  \n",
       "16                  Time since last related instance.  \n",
       "17  Calculates the slope of the linear trend of va...  \n",
       "18  Finds the minimum non-null value of a numeric ...  \n",
       "19  Calculates the number of previous values of an...  \n",
       "20  Creates a transform feature that multplies two...  \n",
       "21           Return the characters in a given string.  \n",
       "22  Calculates the sum of previous values of an in...  \n",
       "23  Creates a transform feature that adds two feat...  \n",
       "24      Compute the time since the previous instance.  \n",
       "25  Transform a Timedelta feature into the number ...  \n",
       "26  Transform Datetime feature into the boolean of...  \n",
       "27  For two boolean values, determine if both valu...  \n",
       "28  Returns the second value on the tuple base fea...  \n",
       "29  Transform a Timedelta feature into the number ...  \n",
       "30  Transform a Timedelta feature into the number ...  \n",
       "31  For each value of the base feature, determines...  \n",
       "32  Compute the difference between the value of a ...  \n",
       "33  Transform a Timedelta feature into the number ...  \n",
       "34  Creates a transform feature that negates a fea...  \n",
       "35  Transform a Timedelta feature into the number ...  \n",
       "36  For each value of the base feature, compute th...  \n",
       "37  Calculates the max of previous values of an in...  \n",
       "38  Creates a transform feature that divides two f...  \n",
       "39         Transform a Datetime feature into the day.  \n",
       "40  Transform a Timedelta feature into the number ...  \n",
       "41  For each value of base feature, return 'True' ...  \n",
       "42  Returns the words in a given string by countin...  \n",
       "43  Calculates the mean of previous values of an i...  \n",
       "44  Creates a transform feature that subtracts two...  \n",
       "45        Transform a Datetime feature into the year.  \n",
       "46  Transform Datetime feature into the boolean of...  \n",
       "47  For two boolean values, determine if one value...  \n",
       "48                    Absolute value of base feature.  \n",
       "49  Calculate the approximate haversine distance i...  \n",
       "50       Transform a Datetime feature into the month.  \n",
       "51  Returns the first value of the tuple base feat...  \n",
       "52        Transform a Datetime feature into the week.  \n",
       "53      Transform a Datetime feature into the minute.  \n",
       "54  For each value of the base feature, negates th...  \n",
       "55             Calculates time since the cutoff time.  \n",
       "56  For each value of the base feature, checks whe...  \n",
       "57      Transform a Datetime feature into the second.  \n",
       "58  Calculates the min of previous values of an in...  \n",
       "59  Creates a transform feature that divides two f...  \n",
       "60  Transform a Timedelta feature into the number ...  \n",
       "61        Transform a Datetime feature into the hour.  "
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ft.list_primitives()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Removed 0 features\n"
     ]
    }
   ],
   "source": [
    "from featuretools import selection\n",
    "\n",
    "# Remove features with only one unique value\n",
    "feature_matrix2 = selection.remove_low_information_features(lala_feature_matrix)\n",
    "\n",
    "print('Removed %d features' % (lala_feature_matrix.shape[1]- feature_matrix2.shape[1]))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(123623, 1426)"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "feature_matrix2.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "df_train_columns = lala_feature_matrix.columns.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100%|██████████| 1426/1426 [00:23<00:00, 59.95it/s]\n"
     ]
    }
   ],
   "source": [
    "from scipy.stats import ks_2samp\n",
    "list_p_value =[]\n",
    "\n",
    "for i in tqdm(df_train_columns):\n",
    "    list_p_value.append(ks_2samp(lala_feature_matrix[i] , lala_feature_matrix[i])[1])\n",
    "\n",
    "Se = pd.Series(list_p_value, index = df_train_columns).sort_values() \n",
    "list_discarded = list(Se[Se < .1].index)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[]"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list_discarded"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "correlation = lala_feature_matrix.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "feather does not support serializing <class 'pandas.core.indexes.base.Index'> for the index; you can .reset_index()to make the index into column(s)",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-17-ab4cfa107305>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[0;32m----> 1\u001b[0;31m \u001b[0mcorrelation\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mto_feather\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0;34m'correlation'\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/anaconda3/envs/fastai/lib/python3.6/site-packages/pandas/core/frame.py\u001b[0m in \u001b[0;36mto_feather\u001b[0;34m(self, fname)\u001b[0m\n\u001b[1;32m   1890\u001b[0m         \"\"\"\n\u001b[1;32m   1891\u001b[0m         \u001b[0;32mfrom\u001b[0m \u001b[0mpandas\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mio\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfeather_format\u001b[0m \u001b[0;32mimport\u001b[0m \u001b[0mto_feather\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 1892\u001b[0;31m         \u001b[0mto_feather\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mself\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mfname\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m   1893\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1894\u001b[0m     def to_parquet(self, fname, engine='auto', compression='snappy',\n",
      "\u001b[0;32m~/anaconda3/envs/fastai/lib/python3.6/site-packages/pandas/io/feather_format.py\u001b[0m in \u001b[0;36mto_feather\u001b[0;34m(df, path)\u001b[0m\n\u001b[1;32m     62\u001b[0m                          \u001b[0;34m\"for the index; you can .reset_index()\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     63\u001b[0m                          \"to make the index into column(s)\".format(\n\u001b[0;32m---> 64\u001b[0;31m                              type(df.index)))\n\u001b[0m\u001b[1;32m     65\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     66\u001b[0m     \u001b[0;32mif\u001b[0m \u001b[0;32mnot\u001b[0m \u001b[0mdf\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mindex\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mequals\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mRangeIndex\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mfrom_range\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mrange\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mlen\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mdf\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mValueError\u001b[0m: feather does not support serializing <class 'pandas.core.indexes.base.Index'> for the index; you can .reset_index()to make the index into column(s)"
     ]
    }
   ],
   "source": [
    "correlation.to_feather('correlation')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "hidden": true
   },
   "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_1</th>\n",
       "      <th>feature_2</th>\n",
       "      <th>feature_3</th>\n",
       "      <th>DAYS_SINCE(first_active_month)</th>\n",
       "      <th>feature_2 + feature_3</th>\n",
       "      <th>feature_1 + feature_2</th>\n",
       "      <th>feature_1 + feature_3</th>\n",
       "      <th>feature_1 % feature_3</th>\n",
       "      <th>feature_1 % feature_2</th>\n",
       "      <th>feature_3 % feature_2</th>\n",
       "      <th>...</th>\n",
       "      <th>feature_1 + feature_2 % feature_1 + feature_3 * feature_2 % feature_2 + feature_3</th>\n",
       "      <th>feature_1 + feature_2 % feature_3 * feature_2 + feature_3 % feature_1 + feature_2</th>\n",
       "      <th>feature_2 + feature_3 % feature_1 + feature_2 * feature_3</th>\n",
       "      <th>feature_1 + feature_2 % feature_1 * feature_3 % feature_2 + feature_3</th>\n",
       "      <th>DAYS_SINCE(first_active_month) % feature_3 * feature_2 + feature_3 % feature_1 + feature_2</th>\n",
       "      <th>feature_1 % feature_1 + feature_2 * feature_1 + feature_2 % feature_1</th>\n",
       "      <th>feature_1 + feature_2 % feature_2 + feature_3 * feature_1 + feature_3 % feature_1</th>\n",
       "      <th>feature_1 + feature_2 % feature_2 + feature_3 * feature_2 % feature_3</th>\n",
       "      <th>feature_2 + feature_3 % feature_1 + feature_3 * feature_2 + feature_3 % feature_2</th>\n",
       "      <th>feature_2 % feature_1 + feature_3 * feature_2 + feature_3</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>feature_1</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.132061</td>\n",
       "      <td>0.582790</td>\n",
       "      <td>0.113843</td>\n",
       "      <td>0.205430</td>\n",
       "      <td>0.826459</td>\n",
       "      <td>0.964794</td>\n",
       "      <td>0.582790</td>\n",
       "      <td>0.038619</td>\n",
       "      <td>-0.250381</td>\n",
       "      <td>...</td>\n",
       "      <td>0.339106</td>\n",
       "      <td>0.291089</td>\n",
       "      <td>0.487026</td>\n",
       "      <td>0.565775</td>\n",
       "      <td>0.262945</td>\n",
       "      <td>0.731561</td>\n",
       "      <td>0.341292</td>\n",
       "      <td>-0.051689</td>\n",
       "      <td>0.097805</td>\n",
       "      <td>0.137869</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>feature_2</th>\n",
       "      <td>-0.132061</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.065344</td>\n",
       "      <td>0.155899</td>\n",
       "      <td>0.844932</td>\n",
       "      <td>0.448923</td>\n",
       "      <td>-0.081353</td>\n",
       "      <td>0.065344</td>\n",
       "      <td>0.448912</td>\n",
       "      <td>0.816185</td>\n",
       "      <td>...</td>\n",
       "      <td>0.018942</td>\n",
       "      <td>0.173283</td>\n",
       "      <td>0.447172</td>\n",
       "      <td>-0.136113</td>\n",
       "      <td>0.162768</td>\n",
       "      <td>-0.088647</td>\n",
       "      <td>0.289003</td>\n",
       "      <td>0.196103</td>\n",
       "      <td>0.282098</td>\n",
       "      <td>0.759679</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>feature_3</th>\n",
       "      <td>0.582790</td>\n",
       "      <td>0.065344</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.195187</td>\n",
       "      <td>0.588942</td>\n",
       "      <td>0.562479</td>\n",
       "      <td>0.775998</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.094626</td>\n",
       "      <td>-0.078709</td>\n",
       "      <td>...</td>\n",
       "      <td>0.288448</td>\n",
       "      <td>0.441731</td>\n",
       "      <td>0.883610</td>\n",
       "      <td>0.529470</td>\n",
       "      <td>0.519518</td>\n",
       "      <td>0.330609</td>\n",
       "      <td>0.413546</td>\n",
       "      <td>0.494299</td>\n",
       "      <td>-0.328349</td>\n",
       "      <td>0.423931</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DAYS_SINCE(first_active_month)</th>\n",
       "      <td>0.113843</td>\n",
       "      <td>0.155899</td>\n",
       "      <td>0.195187</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.230888</td>\n",
       "      <td>0.191172</td>\n",
       "      <td>0.151535</td>\n",
       "      <td>0.195187</td>\n",
       "      <td>-0.194043</td>\n",
       "      <td>0.279974</td>\n",
       "      <td>...</td>\n",
       "      <td>0.188276</td>\n",
       "      <td>-0.105140</td>\n",
       "      <td>0.243634</td>\n",
       "      <td>-0.140469</td>\n",
       "      <td>0.141321</td>\n",
       "      <td>-0.123425</td>\n",
       "      <td>-0.035912</td>\n",
       "      <td>0.219171</td>\n",
       "      <td>-0.053573</td>\n",
       "      <td>0.270002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>feature_2 + feature_3</th>\n",
       "      <td>0.205430</td>\n",
       "      <td>0.844932</td>\n",
       "      <td>0.588942</td>\n",
       "      <td>0.230888</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.665085</td>\n",
       "      <td>0.350062</td>\n",
       "      <td>0.588942</td>\n",
       "      <td>0.414298</td>\n",
       "      <td>0.618844</td>\n",
       "      <td>...</td>\n",
       "      <td>0.169955</td>\n",
       "      <td>0.377119</td>\n",
       "      <td>0.835800</td>\n",
       "      <td>0.173567</td>\n",
       "      <td>0.410299</td>\n",
       "      <td>0.105418</td>\n",
       "      <td>0.455734</td>\n",
       "      <td>0.423779</td>\n",
       "      <td>0.052471</td>\n",
       "      <td>0.842504</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 1426 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                feature_1  feature_2  feature_3  \\\n",
       "feature_1                        1.000000  -0.132061   0.582790   \n",
       "feature_2                       -0.132061   1.000000   0.065344   \n",
       "feature_3                        0.582790   0.065344   1.000000   \n",
       "DAYS_SINCE(first_active_month)   0.113843   0.155899   0.195187   \n",
       "feature_2 + feature_3            0.205430   0.844932   0.588942   \n",
       "\n",
       "                                DAYS_SINCE(first_active_month)  \\\n",
       "feature_1                                             0.113843   \n",
       "feature_2                                             0.155899   \n",
       "feature_3                                             0.195187   \n",
       "DAYS_SINCE(first_active_month)                        1.000000   \n",
       "feature_2 + feature_3                                 0.230888   \n",
       "\n",
       "                                feature_2 + feature_3  feature_1 + feature_2  \\\n",
       "feature_1                                    0.205430               0.826459   \n",
       "feature_2                                    0.844932               0.448923   \n",
       "feature_3                                    0.588942               0.562479   \n",
       "DAYS_SINCE(first_active_month)               0.230888               0.191172   \n",
       "feature_2 + feature_3                        1.000000               0.665085   \n",
       "\n",
       "                                feature_1 + feature_3  feature_1 % feature_3  \\\n",
       "feature_1                                    0.964794               0.582790   \n",
       "feature_2                                   -0.081353               0.065344   \n",
       "feature_3                                    0.775998               1.000000   \n",
       "DAYS_SINCE(first_active_month)               0.151535               0.195187   \n",
       "feature_2 + feature_3                        0.350062               0.588942   \n",
       "\n",
       "                                feature_1 % feature_2  feature_3 % feature_2  \\\n",
       "feature_1                                    0.038619              -0.250381   \n",
       "feature_2                                    0.448912               0.816185   \n",
       "feature_3                                    0.094626              -0.078709   \n",
       "DAYS_SINCE(first_active_month)              -0.194043               0.279974   \n",
       "feature_2 + feature_3                        0.414298               0.618844   \n",
       "\n",
       "                                                          ...                              \\\n",
       "feature_1                                                 ...                               \n",
       "feature_2                                                 ...                               \n",
       "feature_3                                                 ...                               \n",
       "DAYS_SINCE(first_active_month)                            ...                               \n",
       "feature_2 + feature_3                                     ...                               \n",
       "\n",
       "                                feature_1 + feature_2 % feature_1 + feature_3 * feature_2 % feature_2 + feature_3  \\\n",
       "feature_1                                                                0.339106                                   \n",
       "feature_2                                                                0.018942                                   \n",
       "feature_3                                                                0.288448                                   \n",
       "DAYS_SINCE(first_active_month)                                           0.188276                                   \n",
       "feature_2 + feature_3                                                    0.169955                                   \n",
       "\n",
       "                                feature_1 + feature_2 % feature_3 * feature_2 + feature_3 % feature_1 + feature_2  \\\n",
       "feature_1                                                                0.291089                                   \n",
       "feature_2                                                                0.173283                                   \n",
       "feature_3                                                                0.441731                                   \n",
       "DAYS_SINCE(first_active_month)                                          -0.105140                                   \n",
       "feature_2 + feature_3                                                    0.377119                                   \n",
       "\n",
       "                                feature_2 + feature_3 % feature_1 + feature_2 * feature_3  \\\n",
       "feature_1                                                                0.487026           \n",
       "feature_2                                                                0.447172           \n",
       "feature_3                                                                0.883610           \n",
       "DAYS_SINCE(first_active_month)                                           0.243634           \n",
       "feature_2 + feature_3                                                    0.835800           \n",
       "\n",
       "                                feature_1 + feature_2 % feature_1 * feature_3 % feature_2 + feature_3  \\\n",
       "feature_1                                                                0.565775                       \n",
       "feature_2                                                               -0.136113                       \n",
       "feature_3                                                                0.529470                       \n",
       "DAYS_SINCE(first_active_month)                                          -0.140469                       \n",
       "feature_2 + feature_3                                                    0.173567                       \n",
       "\n",
       "                                DAYS_SINCE(first_active_month) % feature_3 * feature_2 + feature_3 % feature_1 + feature_2  \\\n",
       "feature_1                                                                0.262945                                            \n",
       "feature_2                                                                0.162768                                            \n",
       "feature_3                                                                0.519518                                            \n",
       "DAYS_SINCE(first_active_month)                                           0.141321                                            \n",
       "feature_2 + feature_3                                                    0.410299                                            \n",
       "\n",
       "                                feature_1 % feature_1 + feature_2 * feature_1 + feature_2 % feature_1  \\\n",
       "feature_1                                                                0.731561                       \n",
       "feature_2                                                               -0.088647                       \n",
       "feature_3                                                                0.330609                       \n",
       "DAYS_SINCE(first_active_month)                                          -0.123425                       \n",
       "feature_2 + feature_3                                                    0.105418                       \n",
       "\n",
       "                                feature_1 + feature_2 % feature_2 + feature_3 * feature_1 + feature_3 % feature_1  \\\n",
       "feature_1                                                                0.341292                                   \n",
       "feature_2                                                                0.289003                                   \n",
       "feature_3                                                                0.413546                                   \n",
       "DAYS_SINCE(first_active_month)                                          -0.035912                                   \n",
       "feature_2 + feature_3                                                    0.455734                                   \n",
       "\n",
       "                                feature_1 + feature_2 % feature_2 + feature_3 * feature_2 % feature_3  \\\n",
       "feature_1                                                               -0.051689                       \n",
       "feature_2                                                                0.196103                       \n",
       "feature_3                                                                0.494299                       \n",
       "DAYS_SINCE(first_active_month)                                           0.219171                       \n",
       "feature_2 + feature_3                                                    0.423779                       \n",
       "\n",
       "                                feature_2 + feature_3 % feature_1 + feature_3 * feature_2 + feature_3 % feature_2  \\\n",
       "feature_1                                                                0.097805                                   \n",
       "feature_2                                                                0.282098                                   \n",
       "feature_3                                                               -0.328349                                   \n",
       "DAYS_SINCE(first_active_month)                                          -0.053573                                   \n",
       "feature_2 + feature_3                                                    0.052471                                   \n",
       "\n",
       "                                feature_2 % feature_1 + feature_3 * feature_2 + feature_3  \n",
       "feature_1                                                                0.137869          \n",
       "feature_2                                                                0.759679          \n",
       "feature_3                                                                0.423931          \n",
       "DAYS_SINCE(first_active_month)                                           0.270002          \n",
       "feature_2 + feature_3                                                    0.842504          \n",
       "\n",
       "[5 rows x 1426 columns]"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "correlation.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "correlation.index.name = 'Variable'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "hidden": true
   },
   "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_1</th>\n",
       "      <th>feature_2</th>\n",
       "      <th>feature_3</th>\n",
       "      <th>DAYS_SINCE(first_active_month)</th>\n",
       "      <th>feature_2 + feature_3</th>\n",
       "      <th>feature_1 + feature_2</th>\n",
       "      <th>feature_1 + feature_3</th>\n",
       "      <th>feature_1 % feature_3</th>\n",
       "      <th>feature_1 % feature_2</th>\n",
       "      <th>feature_3 % feature_2</th>\n",
       "      <th>...</th>\n",
       "      <th>feature_1 + feature_2 % feature_1 + feature_3 * feature_2 % feature_2 + feature_3</th>\n",
       "      <th>feature_1 + feature_2 % feature_3 * feature_2 + feature_3 % feature_1 + feature_2</th>\n",
       "      <th>feature_2 + feature_3 % feature_1 + feature_2 * feature_3</th>\n",
       "      <th>feature_1 + feature_2 % feature_1 * feature_3 % feature_2 + feature_3</th>\n",
       "      <th>DAYS_SINCE(first_active_month) % feature_3 * feature_2 + feature_3 % feature_1 + feature_2</th>\n",
       "      <th>feature_1 % feature_1 + feature_2 * feature_1 + feature_2 % feature_1</th>\n",
       "      <th>feature_1 + feature_2 % feature_2 + feature_3 * feature_1 + feature_3 % feature_1</th>\n",
       "      <th>feature_1 + feature_2 % feature_2 + feature_3 * feature_2 % feature_3</th>\n",
       "      <th>feature_2 + feature_3 % feature_1 + feature_3 * feature_2 + feature_3 % feature_2</th>\n",
       "      <th>feature_2 % feature_1 + feature_3 * feature_2 + feature_3</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Variable</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>feature_1</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.132061</td>\n",
       "      <td>0.582790</td>\n",
       "      <td>0.113843</td>\n",
       "      <td>0.205430</td>\n",
       "      <td>0.826459</td>\n",
       "      <td>0.964794</td>\n",
       "      <td>0.582790</td>\n",
       "      <td>0.038619</td>\n",
       "      <td>-0.250381</td>\n",
       "      <td>...</td>\n",
       "      <td>0.339106</td>\n",
       "      <td>0.291089</td>\n",
       "      <td>0.487026</td>\n",
       "      <td>0.565775</td>\n",
       "      <td>0.262945</td>\n",
       "      <td>0.731561</td>\n",
       "      <td>0.341292</td>\n",
       "      <td>-0.051689</td>\n",
       "      <td>0.097805</td>\n",
       "      <td>0.137869</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>feature_2</th>\n",
       "      <td>-0.132061</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.065344</td>\n",
       "      <td>0.155899</td>\n",
       "      <td>0.844932</td>\n",
       "      <td>0.448923</td>\n",
       "      <td>-0.081353</td>\n",
       "      <td>0.065344</td>\n",
       "      <td>0.448912</td>\n",
       "      <td>0.816185</td>\n",
       "      <td>...</td>\n",
       "      <td>0.018942</td>\n",
       "      <td>0.173283</td>\n",
       "      <td>0.447172</td>\n",
       "      <td>-0.136113</td>\n",
       "      <td>0.162768</td>\n",
       "      <td>-0.088647</td>\n",
       "      <td>0.289003</td>\n",
       "      <td>0.196103</td>\n",
       "      <td>0.282098</td>\n",
       "      <td>0.759679</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>feature_3</th>\n",
       "      <td>0.582790</td>\n",
       "      <td>0.065344</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.195187</td>\n",
       "      <td>0.588942</td>\n",
       "      <td>0.562479</td>\n",
       "      <td>0.775998</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.094626</td>\n",
       "      <td>-0.078709</td>\n",
       "      <td>...</td>\n",
       "      <td>0.288448</td>\n",
       "      <td>0.441731</td>\n",
       "      <td>0.883610</td>\n",
       "      <td>0.529470</td>\n",
       "      <td>0.519518</td>\n",
       "      <td>0.330609</td>\n",
       "      <td>0.413546</td>\n",
       "      <td>0.494299</td>\n",
       "      <td>-0.328349</td>\n",
       "      <td>0.423931</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>DAYS_SINCE(first_active_month)</th>\n",
       "      <td>0.113843</td>\n",
       "      <td>0.155899</td>\n",
       "      <td>0.195187</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.230888</td>\n",
       "      <td>0.191172</td>\n",
       "      <td>0.151535</td>\n",
       "      <td>0.195187</td>\n",
       "      <td>-0.194043</td>\n",
       "      <td>0.279974</td>\n",
       "      <td>...</td>\n",
       "      <td>0.188276</td>\n",
       "      <td>-0.105140</td>\n",
       "      <td>0.243634</td>\n",
       "      <td>-0.140469</td>\n",
       "      <td>0.141321</td>\n",
       "      <td>-0.123425</td>\n",
       "      <td>-0.035912</td>\n",
       "      <td>0.219171</td>\n",
       "      <td>-0.053573</td>\n",
       "      <td>0.270002</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>feature_2 + feature_3</th>\n",
       "      <td>0.205430</td>\n",
       "      <td>0.844932</td>\n",
       "      <td>0.588942</td>\n",
       "      <td>0.230888</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.665085</td>\n",
       "      <td>0.350062</td>\n",
       "      <td>0.588942</td>\n",
       "      <td>0.414298</td>\n",
       "      <td>0.618844</td>\n",
       "      <td>...</td>\n",
       "      <td>0.169955</td>\n",
       "      <td>0.377119</td>\n",
       "      <td>0.835800</td>\n",
       "      <td>0.173567</td>\n",
       "      <td>0.410299</td>\n",
       "      <td>0.105418</td>\n",
       "      <td>0.455734</td>\n",
       "      <td>0.423779</td>\n",
       "      <td>0.052471</td>\n",
       "      <td>0.842504</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 1426 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                feature_1  feature_2  feature_3  \\\n",
       "Variable                                                          \n",
       "feature_1                        1.000000  -0.132061   0.582790   \n",
       "feature_2                       -0.132061   1.000000   0.065344   \n",
       "feature_3                        0.582790   0.065344   1.000000   \n",
       "DAYS_SINCE(first_active_month)   0.113843   0.155899   0.195187   \n",
       "feature_2 + feature_3            0.205430   0.844932   0.588942   \n",
       "\n",
       "                                DAYS_SINCE(first_active_month)  \\\n",
       "Variable                                                         \n",
       "feature_1                                             0.113843   \n",
       "feature_2                                             0.155899   \n",
       "feature_3                                             0.195187   \n",
       "DAYS_SINCE(first_active_month)                        1.000000   \n",
       "feature_2 + feature_3                                 0.230888   \n",
       "\n",
       "                                feature_2 + feature_3  feature_1 + feature_2  \\\n",
       "Variable                                                                       \n",
       "feature_1                                    0.205430               0.826459   \n",
       "feature_2                                    0.844932               0.448923   \n",
       "feature_3                                    0.588942               0.562479   \n",
       "DAYS_SINCE(first_active_month)               0.230888               0.191172   \n",
       "feature_2 + feature_3                        1.000000               0.665085   \n",
       "\n",
       "                                feature_1 + feature_3  feature_1 % feature_3  \\\n",
       "Variable                                                                       \n",
       "feature_1                                    0.964794               0.582790   \n",
       "feature_2                                   -0.081353               0.065344   \n",
       "feature_3                                    0.775998               1.000000   \n",
       "DAYS_SINCE(first_active_month)               0.151535               0.195187   \n",
       "feature_2 + feature_3                        0.350062               0.588942   \n",
       "\n",
       "                                feature_1 % feature_2  feature_3 % feature_2  \\\n",
       "Variable                                                                       \n",
       "feature_1                                    0.038619              -0.250381   \n",
       "feature_2                                    0.448912               0.816185   \n",
       "feature_3                                    0.094626              -0.078709   \n",
       "DAYS_SINCE(first_active_month)              -0.194043               0.279974   \n",
       "feature_2 + feature_3                        0.414298               0.618844   \n",
       "\n",
       "                                                          ...                              \\\n",
       "Variable                                                  ...                               \n",
       "feature_1                                                 ...                               \n",
       "feature_2                                                 ...                               \n",
       "feature_3                                                 ...                               \n",
       "DAYS_SINCE(first_active_month)                            ...                               \n",
       "feature_2 + feature_3                                     ...                               \n",
       "\n",
       "                                feature_1 + feature_2 % feature_1 + feature_3 * feature_2 % feature_2 + feature_3  \\\n",
       "Variable                                                                                                            \n",
       "feature_1                                                                0.339106                                   \n",
       "feature_2                                                                0.018942                                   \n",
       "feature_3                                                                0.288448                                   \n",
       "DAYS_SINCE(first_active_month)                                           0.188276                                   \n",
       "feature_2 + feature_3                                                    0.169955                                   \n",
       "\n",
       "                                feature_1 + feature_2 % feature_3 * feature_2 + feature_3 % feature_1 + feature_2  \\\n",
       "Variable                                                                                                            \n",
       "feature_1                                                                0.291089                                   \n",
       "feature_2                                                                0.173283                                   \n",
       "feature_3                                                                0.441731                                   \n",
       "DAYS_SINCE(first_active_month)                                          -0.105140                                   \n",
       "feature_2 + feature_3                                                    0.377119                                   \n",
       "\n",
       "                                feature_2 + feature_3 % feature_1 + feature_2 * feature_3  \\\n",
       "Variable                                                                                    \n",
       "feature_1                                                                0.487026           \n",
       "feature_2                                                                0.447172           \n",
       "feature_3                                                                0.883610           \n",
       "DAYS_SINCE(first_active_month)                                           0.243634           \n",
       "feature_2 + feature_3                                                    0.835800           \n",
       "\n",
       "                                feature_1 + feature_2 % feature_1 * feature_3 % feature_2 + feature_3  \\\n",
       "Variable                                                                                                \n",
       "feature_1                                                                0.565775                       \n",
       "feature_2                                                               -0.136113                       \n",
       "feature_3                                                                0.529470                       \n",
       "DAYS_SINCE(first_active_month)                                          -0.140469                       \n",
       "feature_2 + feature_3                                                    0.173567                       \n",
       "\n",
       "                                DAYS_SINCE(first_active_month) % feature_3 * feature_2 + feature_3 % feature_1 + feature_2  \\\n",
       "Variable                                                                                                                     \n",
       "feature_1                                                                0.262945                                            \n",
       "feature_2                                                                0.162768                                            \n",
       "feature_3                                                                0.519518                                            \n",
       "DAYS_SINCE(first_active_month)                                           0.141321                                            \n",
       "feature_2 + feature_3                                                    0.410299                                            \n",
       "\n",
       "                                feature_1 % feature_1 + feature_2 * feature_1 + feature_2 % feature_1  \\\n",
       "Variable                                                                                                \n",
       "feature_1                                                                0.731561                       \n",
       "feature_2                                                               -0.088647                       \n",
       "feature_3                                                                0.330609                       \n",
       "DAYS_SINCE(first_active_month)                                          -0.123425                       \n",
       "feature_2 + feature_3                                                    0.105418                       \n",
       "\n",
       "                                feature_1 + feature_2 % feature_2 + feature_3 * feature_1 + feature_3 % feature_1  \\\n",
       "Variable                                                                                                            \n",
       "feature_1                                                                0.341292                                   \n",
       "feature_2                                                                0.289003                                   \n",
       "feature_3                                                                0.413546                                   \n",
       "DAYS_SINCE(first_active_month)                                          -0.035912                                   \n",
       "feature_2 + feature_3                                                    0.455734                                   \n",
       "\n",
       "                                feature_1 + feature_2 % feature_2 + feature_3 * feature_2 % feature_3  \\\n",
       "Variable                                                                                                \n",
       "feature_1                                                               -0.051689                       \n",
       "feature_2                                                                0.196103                       \n",
       "feature_3                                                                0.494299                       \n",
       "DAYS_SINCE(first_active_month)                                           0.219171                       \n",
       "feature_2 + feature_3                                                    0.423779                       \n",
       "\n",
       "                                feature_2 + feature_3 % feature_1 + feature_3 * feature_2 + feature_3 % feature_2  \\\n",
       "Variable                                                                                                            \n",
       "feature_1                                                                0.097805                                   \n",
       "feature_2                                                                0.282098                                   \n",
       "feature_3                                                               -0.328349                                   \n",
       "DAYS_SINCE(first_active_month)                                          -0.053573                                   \n",
       "feature_2 + feature_3                                                    0.052471                                   \n",
       "\n",
       "                                feature_2 % feature_1 + feature_3 * feature_2 + feature_3  \n",
       "Variable                                                                                   \n",
       "feature_1                                                                0.137869          \n",
       "feature_2                                                                0.759679          \n",
       "feature_3                                                                0.423931          \n",
       "DAYS_SINCE(first_active_month)                                           0.270002          \n",
       "feature_2 + feature_3                                                    0.842504          \n",
       "\n",
       "[5 rows x 1426 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "correlation.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "correlation.reset_index().to_feather('correlation')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "threshold = 0.9\n",
    "\n",
    "correlated_pairs = {}\n",
    "\n",
    "# Iterate through the columns\n",
    "for col in correlation:\n",
    "    # Find correlations above the threshold\n",
    "    above_threshold_vars = [x for x in list(correlation.index[correlation[col] > threshold]) if x != col]\n",
    "    correlated_pairs[col] = above_threshold_vars"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Variable\n",
       "DAYS_SINCE(first_active_month) + feature_1 % DAYS_SINCE(first_active_month) + feature_2         -0.322750\n",
       "DAYS_SINCE(first_active_month) + feature_3 % DAYS_SINCE(first_active_month) + feature_2         -0.280467\n",
       "feature_3 % feature_2                                                                           -0.250381\n",
       "feature_2 + feature_3 % feature_2                                                               -0.250381\n",
       "feature_1 % feature_2 * feature_3 % feature_2                                                   -0.236594\n",
       "feature_1 + feature_2 % feature_2 * feature_3 % feature_2                                       -0.236594\n",
       "feature_1 + feature_2 % feature_2 * feature_2 + feature_3 % feature_2                           -0.236594\n",
       "feature_1 % feature_2 * feature_2 + feature_3 % feature_2                                       -0.236594\n",
       "feature_1 + feature_2 % feature_1 + feature_3 * feature_3 % feature_2                           -0.220574\n",
       "feature_1 + feature_2 % feature_1 + feature_3 * feature_2 + feature_3 % feature_2               -0.220574\n",
       "feature_2 * feature_2 + feature_3 % feature_2                                                   -0.206110\n",
       "feature_2 * feature_3 % feature_2                                                               -0.206110\n",
       "feature_2 % feature_2 + feature_3 * feature_2 + feature_3 % feature_2                           -0.206110\n",
       "feature_2 % feature_1 + feature_2 * feature_3 % feature_2                                       -0.206110\n",
       "feature_2 % feature_2 + feature_3 * feature_3 % feature_2                                       -0.206110\n",
       "feature_2 % DAYS_SINCE(first_active_month) * feature_3 % feature_2                              -0.206110\n",
       "feature_2 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_2                  -0.206110\n",
       "feature_2 % feature_1 + feature_2 * feature_2 + feature_3 % feature_2                           -0.206110\n",
       "feature_1 % feature_2 + feature_3 * feature_2 + feature_3 % feature_2                           -0.193307\n",
       "feature_1 % feature_2 + feature_3 * feature_3 % feature_2                                       -0.193307\n",
       "feature_2 + feature_3 * feature_3 % feature_2                                                   -0.186529\n",
       "feature_2 + feature_3 * feature_2 + feature_3 % feature_2                                       -0.186529\n",
       "feature_2 + feature_3 % DAYS_SINCE(first_active_month) * feature_3 % feature_2                  -0.186529\n",
       "feature_2 + feature_3 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_2      -0.186529\n",
       "DAYS_SINCE(first_active_month) + feature_2 * feature_3 % feature_2                              -0.185294\n",
       "DAYS_SINCE(first_active_month) + feature_2 * feature_2 + feature_3 % feature_2                  -0.185294\n",
       "DAYS_SINCE(first_active_month) * feature_3 % feature_2                                          -0.185178\n",
       "DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_2                              -0.185178\n",
       "DAYS_SINCE(first_active_month) + feature_3 * feature_2 + feature_3 % feature_2                  -0.185161\n",
       "DAYS_SINCE(first_active_month) + feature_3 * feature_3 % feature_2                              -0.185161\n",
       "                                                                                                   ...   \n",
       "feature_1 + feature_3 * feature_2 + feature_3 % feature_2                                       -0.134350\n",
       "feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_3 % feature_2                  -0.134350\n",
       "feature_1 + feature_3 % feature_1 + feature_2 * feature_2 + feature_3 % feature_2               -0.134350\n",
       "feature_1 + feature_3 % feature_1 + feature_2 * feature_3 % feature_2                           -0.134350\n",
       "feature_1 + feature_3 * feature_3 % feature_2                                                   -0.134350\n",
       "DAYS_SINCE(first_active_month) % feature_2 * feature_1 + feature_2 % feature_1 + feature_3      -0.134263\n",
       "feature_2 % feature_1 + feature_2                                                               -0.132061\n",
       "feature_2 % feature_2 + feature_3                                                               -0.132061\n",
       "DAYS_SINCE(first_active_month) + feature_2 % DAYS_SINCE(first_active_month)                     -0.132061\n",
       "feature_2 % DAYS_SINCE(first_active_month) + feature_2                                          -0.132061\n",
       "feature_2 % DAYS_SINCE(first_active_month) + feature_1                                          -0.132061\n",
       "feature_2 % DAYS_SINCE(first_active_month)                                                      -0.132061\n",
       "feature_2 % DAYS_SINCE(first_active_month) + feature_3                                          -0.132061\n",
       "feature_2                                                                                       -0.132061\n",
       "feature_1 % feature_2 + feature_3 * feature_1 + feature_3 % feature_1 + feature_2               -0.125470\n",
       "DAYS_SINCE(first_active_month) % feature_2 * feature_3 % feature_2                              -0.123493\n",
       "DAYS_SINCE(first_active_month) % feature_2 * feature_2 + feature_3 % feature_2                  -0.123493\n",
       "feature_1 % DAYS_SINCE(first_active_month) * feature_3 % feature_2                              -0.120095\n",
       "feature_1 % feature_1 + feature_3 * feature_3 % feature_2                                       -0.120095\n",
       "feature_1 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_2                  -0.120095\n",
       "feature_1 % feature_1 + feature_2 * feature_2 + feature_3 % feature_2                           -0.120095\n",
       "feature_1 % feature_1 + feature_2 * feature_3 % feature_2                                       -0.120095\n",
       "feature_1 * feature_2 + feature_3 % feature_2                                                   -0.120095\n",
       "feature_1 % feature_1 + feature_3 * feature_2 + feature_3 % feature_2                           -0.120095\n",
       "feature_1 * feature_3 % feature_2                                                               -0.120095\n",
       "DAYS_SINCE(first_active_month) % feature_2 * feature_1 % feature_2 + feature_3                  -0.117696\n",
       "feature_1 + feature_2 % feature_2 + feature_3 * feature_1 + feature_3 % feature_1 + feature_2   -0.116820\n",
       "feature_2 + feature_3 % feature_2 * feature_3 % feature_1                                       -0.116203\n",
       "feature_3 % feature_1 * feature_3 % feature_2                                                   -0.116203\n",
       "feature_1 + feature_3 % feature_1 * feature_3 % feature_2                                       -0.116203\n",
       "Name: feature_1, Length: 100, dtype: float64"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "correlation['feature_1'].sort_values(ascending=True)[:100]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "corr_matrix = correlation.abs()\n",
    "\n",
    "# Select upper triangle of correlation matrix\n",
    "upper = corr_matrix.where(np.triu(np.ones(corr_matrix.shape), k=1).astype(np.bool))\n",
    "\n",
    "# Find features with correlation greater than 0.95\n",
    "to_drop = [column for column in upper.columns if any(upper[column] > 0.95)]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "1189"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(to_drop)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "hidden": true,
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['feature_1 + feature_3',\n",
       " 'feature_1 % feature_3',\n",
       " 'feature_1 * feature_3',\n",
       " 'feature_2 * feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2',\n",
       " 'feature_2 + feature_3 % feature_2',\n",
       " 'feature_1 + feature_2 % feature_2',\n",
       " 'feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 % feature_1 + feature_2',\n",
       " 'feature_1 % feature_1 + feature_3',\n",
       " 'feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 % feature_3',\n",
       " 'feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_1',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 + feature_3 % feature_3',\n",
       " 'feature_3 % feature_1 + feature_3',\n",
       " 'feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % feature_1',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % feature_3 * feature_3 % feature_1',\n",
       " 'feature_2 % feature_1 * feature_2 + feature_3',\n",
       " 'feature_1 % feature_3 * feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 * feature_3',\n",
       " 'feature_2 + feature_3 * feature_3 % feature_2',\n",
       " 'feature_2 * feature_2 + feature_3',\n",
       " 'feature_1 % feature_3 * feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) * feature_3',\n",
       " 'feature_1 % feature_3 * feature_2',\n",
       " 'feature_2 * feature_3 % feature_1',\n",
       " 'feature_2 + feature_3 * feature_3 % feature_1',\n",
       " 'feature_1 * feature_1 % feature_3',\n",
       " 'feature_1 * feature_1 + feature_3',\n",
       " 'feature_1 % feature_3 * feature_3',\n",
       " 'feature_1 + feature_3 * feature_3 % feature_2',\n",
       " 'feature_1 + feature_3 * feature_2 % feature_1',\n",
       " 'feature_2 + feature_3 * feature_3',\n",
       " 'feature_1 % feature_2 * feature_1 % feature_3',\n",
       " 'feature_1 + feature_3 * feature_2',\n",
       " 'feature_1 % feature_2 * feature_2',\n",
       " 'feature_1 % feature_3 * feature_2 % feature_3',\n",
       " 'feature_1 + feature_3 * feature_2 % feature_3',\n",
       " 'feature_1 + feature_2 * feature_2 % feature_3',\n",
       " 'feature_1 * feature_3 % feature_2',\n",
       " 'feature_1 + feature_3 * feature_3 % feature_1',\n",
       " 'feature_1 % feature_3 * feature_3 % feature_1',\n",
       " 'feature_1 % feature_2 * feature_1 + feature_3',\n",
       " 'feature_1 * feature_3 % feature_1',\n",
       " 'feature_1 + feature_2 * feature_1 + feature_3',\n",
       " 'feature_2 * feature_3 % feature_2',\n",
       " 'feature_1 % feature_2 * feature_3',\n",
       " 'feature_3 * feature_3 % feature_2',\n",
       " 'feature_1 + feature_2 * feature_3 % feature_1',\n",
       " 'feature_1 % feature_2 * feature_3 % feature_1',\n",
       " 'feature_1 * feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) * feature_1',\n",
       " 'feature_2 % feature_3 * feature_3',\n",
       " 'feature_2 * feature_2 % feature_1',\n",
       " 'feature_1 + feature_2 * feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 * feature_2 % feature_1',\n",
       " 'feature_2 % feature_3 * feature_3 % feature_2',\n",
       " 'feature_1 * feature_1 + feature_2',\n",
       " 'feature_1 * feature_1 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) * feature_1 % feature_2',\n",
       " 'feature_1 * feature_2 % feature_3',\n",
       " 'feature_2 % feature_1 * feature_3 % feature_1',\n",
       " 'feature_1 + feature_2 * feature_3',\n",
       " 'feature_1 % feature_2 * feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 * feature_3 % feature_2',\n",
       " 'feature_2 % feature_3 * feature_2 + feature_3',\n",
       " 'feature_3 * feature_3 % feature_1',\n",
       " 'feature_1 + feature_3 * feature_2 + feature_3',\n",
       " 'feature_1 % feature_3 * feature_1 + feature_2',\n",
       " 'feature_3 % feature_1 * feature_3 % feature_2',\n",
       " 'feature_1 % feature_2 * feature_1 + feature_2',\n",
       " 'feature_3 % DAYS_SINCE(first_active_month) + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 % feature_3',\n",
       " 'feature_3 % DAYS_SINCE(first_active_month) + feature_1',\n",
       " 'feature_3 % DAYS_SINCE(first_active_month) + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % DAYS_SINCE(first_active_month) + feature_2',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 % feature_2',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) + feature_1',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) + feature_3',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) + feature_2',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) + feature_1',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) + feature_2',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) + feature_1',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) + feature_2',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) + feature_3',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) + feature_1',\n",
       " 'feature_2 + feature_3 % DAYS_SINCE(first_active_month) + feature_1',\n",
       " 'feature_2 + feature_3 % DAYS_SINCE(first_active_month) + feature_2',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) + feature_3',\n",
       " 'feature_2 + feature_3 % DAYS_SINCE(first_active_month) + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % DAYS_SINCE(first_active_month) + feature_1',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % DAYS_SINCE(first_active_month) + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_3 % feature_1',\n",
       " 'feature_1 % feature_3 * feature_3 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_1 + feature_2 % feature_1',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_3 % feature_1',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 + feature_3 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_2 + feature_3 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_2',\n",
       " 'feature_1 + feature_3 % feature_1 + feature_2 * feature_2 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_1 % feature_1 + feature_3',\n",
       " 'feature_2 + feature_3 % feature_1 + feature_3 * feature_3 % feature_1',\n",
       " 'feature_1 + feature_2 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_3',\n",
       " 'feature_1 % feature_3 * feature_2 % feature_1 + feature_3',\n",
       " 'feature_2 % feature_1 + feature_3 * feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % feature_1 + feature_2 * feature_3',\n",
       " 'feature_1 + feature_3 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 * feature_2 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 * feature_3',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 % feature_1 + feature_2 * feature_2 % feature_1 + feature_2',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) * feature_2 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_2 + feature_3 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_2 % feature_2 + feature_3',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 * feature_2 + feature_3 % feature_2',\n",
       " 'feature_1 + feature_3 % feature_1 + feature_2 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 + feature_3 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % feature_2 * feature_2 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 % feature_3 * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % feature_3 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_3 % feature_1 + feature_2 * feature_3 % feature_2',\n",
       " 'feature_1 + feature_2 * feature_1 + feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 * feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_3 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_2 % feature_3 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_2 % feature_1 + feature_2 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % feature_1 + feature_2 * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_1 + feature_2',\n",
       " 'feature_1 + feature_2 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_3 * feature_1 + feature_2 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) * feature_2 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 + feature_3 * feature_2 + feature_3 % feature_2',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_3 % feature_2 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_2',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_2 % feature_1',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_1 + feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_2 * feature_1 + feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % feature_2 + feature_3 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 * feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_2 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % feature_1 + feature_2 * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_3',\n",
       " 'feature_1 % feature_2 * feature_1 + feature_2 % feature_2',\n",
       " 'feature_1 + feature_2 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_1 + feature_2 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 * feature_1 + feature_2 % feature_2',\n",
       " 'feature_1 + feature_2 % feature_1 + feature_3 * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % feature_2 * feature_2 + feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_1 + feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_1 % feature_2',\n",
       " 'feature_2 + feature_3 % feature_2 * feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_2 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 * feature_2 % feature_1 + feature_3',\n",
       " 'feature_2 + feature_3 * feature_2 + feature_3 % feature_1',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_2 + feature_3 % feature_2 * feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 + feature_3 % DAYS_SINCE(first_active_month) * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_1 + feature_3 % feature_2',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_1 + feature_2 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_1 % feature_1 + feature_2',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_2',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_3',\n",
       " 'feature_1 + feature_3 % feature_2 + feature_3 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_2 % feature_1 + feature_3 * feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_1 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_2 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * DAYS_SINCE(first_active_month) + feature_3',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 + feature_3 % feature_3 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 % feature_3 * feature_2 + feature_3 % feature_2',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_1 + feature_3 % feature_1',\n",
       " 'feature_2 + feature_3 % feature_1 * feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) * feature_3 % feature_2 + feature_3',\n",
       " 'feature_2 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % feature_1 + feature_3 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 + feature_3 % feature_2 * feature_3 % feature_1',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_2 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_3',\n",
       " 'feature_2 % feature_2 + feature_3 * feature_3 % feature_2',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_2 % feature_2 + feature_3 * feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_1 + feature_2 % feature_1',\n",
       " 'feature_1 % feature_2 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_2 + feature_3 % feature_1 + feature_2 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 * feature_1 + feature_3 % feature_1',\n",
       " 'feature_1 + feature_2 % feature_2 + feature_3 * feature_2 % feature_1 + feature_3',\n",
       " 'feature_2 + feature_3 % feature_2 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_2 % feature_3',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_2 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_2 + feature_3',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) * feature_2 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 * feature_1 % feature_3',\n",
       " 'feature_1 + feature_3 % feature_1 + feature_2 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % feature_1 + feature_3 * feature_1 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_2 * feature_1 + feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * DAYS_SINCE(first_active_month) + feature_1',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_2 % feature_1 + feature_3 * feature_1 + feature_3 % feature_1',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_2 + feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % feature_1 * feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_1 + feature_3',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_2 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_3 % feature_2',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_1 + feature_3',\n",
       " 'feature_2 % feature_2 + feature_3 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_3 % DAYS_SINCE(first_active_month) * feature_3 % feature_2',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_1 + feature_3 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * DAYS_SINCE(first_active_month) + feature_3',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_1 + feature_3 % feature_3',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_1 + feature_2 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_3 % feature_2',\n",
       " 'feature_1 + feature_2 % feature_1 + feature_3 * feature_3',\n",
       " 'feature_2 % feature_1 * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_2 % feature_3',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_3 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_2 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_2 + feature_3 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_1 % feature_1 + feature_3',\n",
       " 'feature_1 * feature_1 + feature_2 % feature_2',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_2 % feature_2 + feature_3',\n",
       " 'feature_3 % DAYS_SINCE(first_active_month) * feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_2 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_1 + feature_3',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_2 % feature_1 + feature_2',\n",
       " 'feature_1 * feature_1 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_2 % feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_3 % feature_2',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_1 + feature_3 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_2 % feature_1 + feature_2',\n",
       " 'feature_2 + feature_3 % feature_1 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_1 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_2 % feature_1',\n",
       " 'feature_2 * feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_1 + feature_2 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_2 + feature_3 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_1',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_1 + feature_3 % feature_2',\n",
       " 'feature_1 + feature_2 * feature_1 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_1 + feature_2 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_1 % feature_3',\n",
       " 'feature_1 + feature_3 * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 + feature_3 % feature_1 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_1 + feature_2',\n",
       " 'feature_1 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_1 + feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 * feature_2 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 * feature_1 + feature_2 % feature_3',\n",
       " 'feature_2 + feature_3 % feature_1 + feature_2 * feature_3 % feature_1',\n",
       " 'feature_1 % feature_3 * feature_1 + feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_1',\n",
       " 'feature_1 + feature_3 % feature_2 * feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_1',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_3 % feature_2',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_3',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 % feature_3 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 * feature_1 + feature_3 % feature_1',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_2 % feature_1',\n",
       " 'feature_2 % feature_3 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_1 + feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_2 + feature_3 % feature_1',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_2 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_1 + feature_2 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) * feature_1 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_2 % feature_3',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 * feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_1 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_3 % feature_1 + feature_2 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % feature_2 + feature_3 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_2 % feature_2 + feature_3 * feature_2 + feature_3 % feature_2',\n",
       " 'feature_2 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 + feature_3 % feature_1 + feature_3 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_1 % feature_2',\n",
       " 'feature_2 * feature_2 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 % feature_2 + feature_3 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_1',\n",
       " 'feature_3 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_2 + feature_3 % feature_3 * feature_3',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 % feature_3 * feature_2 % feature_2 + feature_3',\n",
       " 'feature_2 * feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_1 % feature_3',\n",
       " 'feature_1 + feature_2 * feature_1 + feature_3 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_2 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_2 % feature_1',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_2 % feature_3',\n",
       " 'feature_1 + feature_3 % feature_1 + feature_2 * feature_2 + feature_3 % feature_2',\n",
       " 'feature_3 % DAYS_SINCE(first_active_month) * feature_3 % feature_1',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_2 + feature_3',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_1 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_2 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_2 + feature_3 % feature_1',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_2 % feature_1 + feature_3',\n",
       " 'feature_1 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * DAYS_SINCE(first_active_month) + feature_2',\n",
       " 'feature_1 * feature_2 + feature_3 % feature_1',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_2 + feature_3 % feature_1',\n",
       " 'feature_1 + feature_2 % feature_2 + feature_3 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_1 + feature_3 % feature_2',\n",
       " 'feature_1 + feature_3 % feature_2 + feature_3 * feature_2 % feature_1',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_2 % feature_3',\n",
       " 'feature_1 * feature_1 + feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_1 + feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 + feature_3 % feature_1 + feature_2 * feature_2 + feature_3 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_1',\n",
       " 'feature_1 + feature_3 % feature_1 + feature_2 * feature_2 % feature_1',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_1 + feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_1 + feature_2 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_2 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 * feature_1 % feature_1 + feature_2',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_1 + feature_3 % feature_2',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_2',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_2 + feature_3',\n",
       " 'feature_1 + feature_3 * feature_1 + feature_3 % feature_3',\n",
       " 'feature_1 * feature_2 + feature_3 % feature_2',\n",
       " 'feature_1 + feature_2 % feature_1 + feature_3 * feature_2 + feature_3 % feature_2',\n",
       " 'feature_1 % feature_3 * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 % feature_3 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_1 + feature_3 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_1 + feature_3 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_2',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_3 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_2',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_1 + feature_2 % feature_3',\n",
       " 'feature_1 * feature_2 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_1 + feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_1 + feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) * feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_1 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_2 + feature_3 * feature_2 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_1 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % feature_1 + feature_3 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % feature_2 + feature_3 * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_1 + feature_3',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_1 + feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 % feature_2 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_2 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_1 + feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_2 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_1 + feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % feature_1 + feature_3 * feature_1 + feature_2 % feature_2',\n",
       " 'feature_1 % feature_3 * feature_1 + feature_2 % feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_2 + feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_2 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_3',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_2 % feature_3',\n",
       " 'feature_2 % feature_1 + feature_2 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 * feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_2 + feature_3 % feature_1',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_2 % feature_1 + feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) * feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_1 % feature_2',\n",
       " 'feature_1 % feature_2 * feature_1 + feature_2 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_3 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_2 + feature_3 % feature_2',\n",
       " 'feature_1 + feature_2 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_1 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_3 % feature_1 + feature_2 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_1 + feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_1 + feature_2 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_1 + feature_3 % feature_1',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) * feature_3',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_2 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * DAYS_SINCE(first_active_month) + feature_2',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) * feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_1 % feature_3',\n",
       " 'feature_1 * feature_1 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_1 + feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_3 % feature_1',\n",
       " 'feature_3 % feature_2 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_3 * feature_1 + feature_3 % feature_2 + feature_3',\n",
       " 'feature_2 + feature_3 * feature_3 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_2 + feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_1 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_3 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_1 % feature_1 + feature_2',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_2 % feature_1',\n",
       " 'feature_1 + feature_2 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_2',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % feature_1 * feature_3 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) * feature_1 + feature_3 % feature_1',\n",
       " 'feature_2 % feature_1 + feature_2 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_2 % feature_1 + feature_2',\n",
       " 'feature_1 * feature_1 + feature_3 % feature_1',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_1 + feature_2 % feature_1 + feature_3',\n",
       " 'feature_1 % feature_2 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) * feature_1 + feature_2 % feature_3',\n",
       " 'feature_2 % feature_2 + feature_3 * feature_2 % feature_3',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_2 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_2 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_1 + feature_2 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_1 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) * DAYS_SINCE(first_active_month) % feature_3',\n",
       " 'feature_2 % feature_1 + feature_2 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_3',\n",
       " 'feature_1 + feature_3 % feature_2 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_1 + feature_3',\n",
       " 'feature_2 % feature_1 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 * feature_1 + feature_3 % feature_3',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_1 + feature_3 % feature_3',\n",
       " 'feature_2 + feature_3 % feature_3 * feature_3 % feature_2',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_2 % feature_1',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_2 + feature_3 % feature_1',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_1 + feature_2 % feature_3',\n",
       " 'feature_2 + feature_3 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_2 % feature_2 + feature_3 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_1 + feature_3 % feature_1',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) * feature_3 % feature_2 + feature_3',\n",
       " 'feature_2 + feature_3 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_1 % feature_1 + feature_2',\n",
       " 'feature_2 + feature_3 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) * feature_1 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_2 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_1 + feature_2 * feature_1 + feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * DAYS_SINCE(first_active_month) + feature_3',\n",
       " 'feature_1 + feature_3 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % feature_2 + feature_3 * feature_1 + feature_3 % feature_2',\n",
       " 'feature_2 + feature_3 * feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * DAYS_SINCE(first_active_month) + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_1 + feature_3 % feature_2',\n",
       " 'feature_2 + feature_3 % DAYS_SINCE(first_active_month) * feature_3 % feature_2 + feature_3',\n",
       " 'feature_2 % feature_1 + feature_3 * feature_3 % feature_1',\n",
       " 'feature_2 + feature_3 % feature_3 * feature_3 % feature_1',\n",
       " 'feature_1 % feature_3 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_2 + feature_3',\n",
       " 'feature_2 + feature_3 % feature_2 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 * feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_2 % feature_1',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_3',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_1 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_3 % DAYS_SINCE(first_active_month) * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_1 + feature_2 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_3',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_1 + feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_2 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 * feature_1 + feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % feature_1 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_1 + feature_2 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) * feature_3 % feature_1 + feature_2',\n",
       " 'feature_2 + feature_3 % feature_3 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 * feature_2 + feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * DAYS_SINCE(first_active_month) + feature_1',\n",
       " 'feature_1 + feature_3 % feature_1 + feature_2 * feature_3 % feature_1',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_1 + feature_3 % feature_1',\n",
       " 'feature_3 % DAYS_SINCE(first_active_month) * feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_3 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 + feature_3 % feature_3 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 * feature_1 + feature_3 % feature_2',\n",
       " 'feature_2 % feature_1 + feature_3 * feature_2 + feature_3 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_1 + feature_3 % feature_1',\n",
       " 'feature_2 % feature_1 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_3 % feature_1',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_2 + feature_3',\n",
       " 'feature_1 % feature_2 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_1 + feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) * feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % feature_1 + feature_2 * feature_3 % feature_1',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 + feature_3 % feature_1 + feature_2 * feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_3 % feature_1',\n",
       " 'feature_2 % feature_1 + feature_3 * feature_2 + feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_1 + feature_2',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_2 + feature_3 % feature_2',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_2 % feature_1 + feature_3',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) * feature_2 % feature_1 + feature_2',\n",
       " 'feature_2 + feature_3 % feature_1 + feature_2 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_2 + feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 % feature_2 * feature_1 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_2',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * DAYS_SINCE(first_active_month) + feature_1',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_1 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_2 + feature_3 * feature_3',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_2',\n",
       " 'feature_1 + feature_2 * feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) * DAYS_SINCE(first_active_month) + feature_3',\n",
       " 'feature_1 + feature_3 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_2 % feature_1 + feature_2 * feature_3 % feature_2',\n",
       " 'feature_3 % feature_1 + feature_3 * feature_3 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_1 + feature_3 % feature_2',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_2 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_1 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_2 + feature_3 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_1 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_1 % feature_1 + feature_3',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_2 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_1 + feature_2 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_1',\n",
       " 'feature_2 % feature_1 + feature_2 * feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_2 % feature_3',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_1 + feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_2 % feature_3',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_2 % feature_1 + feature_3 * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_3 % feature_2 * feature_2 % feature_3',\n",
       " 'feature_2 + feature_3 % feature_1 + feature_3 * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_2 % feature_1 + feature_3 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_2 + feature_3 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_1',\n",
       " 'feature_2 + feature_3 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_2 * feature_2 + feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_1 + feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_1 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_2 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_2 % feature_1',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_1 + feature_2 % feature_2',\n",
       " 'feature_1 + feature_2 % feature_1 + feature_3 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_3 * feature_2 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_1 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_3',\n",
       " 'feature_2 + feature_3 % feature_2 * feature_3 % feature_2',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_1 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_1',\n",
       " 'feature_1 + feature_2 % feature_1 + feature_3 * feature_2 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_1 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_2 % feature_2 + feature_3 * feature_2 % feature_1',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_1 + feature_3 % feature_3',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_3 % feature_2',\n",
       " 'feature_1 + feature_2 * feature_2 + feature_3 % feature_1',\n",
       " 'feature_1 + feature_3 % feature_2 + feature_3 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_1 + feature_3 * feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_1 + feature_2',\n",
       " 'feature_3 % feature_1 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_2 % feature_1 + feature_2 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_1 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_2 % feature_1 + feature_2',\n",
       " 'feature_2 % feature_1 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_2 * feature_2 + feature_3 % feature_1',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_3 % feature_1 + feature_3 * feature_3 % feature_2',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_2 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_2',\n",
       " 'feature_1 + feature_3 * feature_2 % feature_1 + feature_2',\n",
       " 'feature_1 % feature_2 * feature_1 + feature_2 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 * feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_2',\n",
       " 'feature_1 + feature_2 % feature_1 + feature_3 * feature_3 % feature_2',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_1 + feature_3 % feature_3',\n",
       " 'feature_2 % feature_2 + feature_3 * feature_3 % feature_1',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_2',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_1',\n",
       " 'feature_1 + feature_3 % feature_1 + feature_2 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_1 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 * feature_1 + feature_2 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_3',\n",
       " 'feature_1 * feature_2 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_2 + feature_3',\n",
       " 'feature_1 % feature_3 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_2 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_1 + feature_3 % feature_3',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_1 + feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % feature_1 + feature_2 * feature_1 + feature_3 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_1 + feature_3 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_1 + feature_2 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_1 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_2 + feature_3',\n",
       " 'feature_2 % feature_2 + feature_3 * feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) * feature_1 + feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 * feature_1 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_1 + feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % feature_3 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_3 * feature_1 + feature_2 % feature_1',\n",
       " 'feature_2 % feature_1 + feature_3 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_2 + feature_3 % feature_3 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_3 * feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) * feature_1 + feature_2 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_2 % feature_1',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_2 % feature_1 + feature_2',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_3 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_3 % feature_2',\n",
       " 'feature_1 % feature_3 * feature_1 + feature_3 % feature_1',\n",
       " 'feature_2 + feature_3 * feature_2 + feature_3 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) * feature_1 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_2 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 * feature_1 + feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_1 + feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 + feature_3 % feature_1 * feature_2 + feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_3 % feature_1',\n",
       " 'feature_2 * feature_2 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_2 * feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) * feature_1 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_2 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_3 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 * feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 * feature_1 + feature_2 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_1 + feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_2 + feature_3 % feature_2',\n",
       " 'feature_1 % feature_2 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_2 + feature_3 % DAYS_SINCE(first_active_month) * feature_3 % feature_2',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) * feature_2 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_1 + feature_2 % feature_3',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_1',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_1 + feature_2 % feature_1',\n",
       " 'feature_2 % feature_2 + feature_3 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_2 % feature_1',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) * feature_3 % feature_2',\n",
       " 'feature_1 % feature_3 * feature_1 + feature_2 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_2 + feature_3 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_1 + feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_2 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_1 + feature_3 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_1 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_1 + feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_2 * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_2 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_2 % feature_2 + feature_3 * feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_2 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) * DAYS_SINCE(first_active_month) % feature_1 + feature_2',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_2 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_1 % feature_2 + feature_3',\n",
       " 'feature_2 + feature_3 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_2 % feature_1 + feature_3 * feature_2 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_1 + feature_2 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_1 % feature_1 + feature_2',\n",
       " 'feature_1 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_2',\n",
       " 'feature_1 + feature_2 % feature_1 + feature_3 * feature_2 + feature_3 % feature_1',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_2 + feature_3',\n",
       " 'feature_2 + feature_3 % DAYS_SINCE(first_active_month) * feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_2 % feature_2 + feature_3 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) * feature_2 % feature_1',\n",
       " 'feature_1 * feature_1 + feature_2 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 % feature_2 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_1 + feature_2 % feature_2',\n",
       " 'feature_1 % feature_2 * feature_2 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 * feature_1 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_3 % feature_1',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 * feature_1 + feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_1 + feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_3 % feature_2 + feature_3',\n",
       " 'feature_2 + feature_3 % feature_1 * feature_3 % feature_2',\n",
       " 'feature_2 % feature_1 + feature_3 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_1 + feature_3 % feature_1',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_1 + feature_2 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 + feature_3 % feature_1 + feature_3 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_2 * feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_1 + feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_1 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_2 + feature_3 * feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * DAYS_SINCE(first_active_month) + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_1 + feature_3 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_1 + feature_2 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_1 + feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 * feature_1',\n",
       " 'feature_1 + feature_2 % feature_2 + feature_3 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 % feature_1 + feature_2 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_2 + feature_3 % feature_1 + feature_2 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_2 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_1 + feature_2 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_1 + feature_2 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) * feature_1 % feature_1 + feature_2',\n",
       " 'feature_1 % feature_3 * feature_1 + feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_1 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % feature_2 + feature_3 * feature_2 % feature_1 + feature_2',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_1 + feature_2 % feature_2',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_2 % feature_1 + feature_2',\n",
       " 'feature_1 % feature_3 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % feature_1 * feature_2 % feature_1 + feature_2',\n",
       " 'feature_2 % feature_1 + feature_3 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_2 + feature_3 % feature_1 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_2 % feature_1 * feature_1 + feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) * DAYS_SINCE(first_active_month) % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_3 % feature_2',\n",
       " 'feature_1 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_2 + feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_1 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % feature_2 * feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_1 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_1 + feature_3 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_2',\n",
       " 'feature_2 % feature_3 * feature_2 + feature_3 % feature_2',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % feature_1 + feature_2 * feature_3',\n",
       " 'DAYS_SINCE(first_active_month) * DAYS_SINCE(first_active_month) % feature_1',\n",
       " 'feature_1 + feature_3 % feature_1 + feature_2 * feature_2 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 * feature_1 + feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_2 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 + feature_3 % feature_1 + feature_2 * feature_3 % feature_2',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_2',\n",
       " 'feature_2 % feature_1 * feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_2 % feature_3 * feature_3 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) * DAYS_SINCE(first_active_month) + feature_1',\n",
       " 'feature_1 + feature_3 % feature_2 * feature_2 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_1 + feature_2 % feature_2',\n",
       " 'feature_2 + feature_3 % feature_2 * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 % feature_2 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_2 + feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_2 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_1 + feature_2 * feature_2 % feature_2 + feature_3',\n",
       " 'feature_2 % feature_1 + feature_2 * feature_2 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_1 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_3',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_1 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 * feature_2 + feature_3 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 % feature_3 * feature_1 + feature_3 % feature_3',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_2 % feature_3 * feature_2 + feature_3 % feature_1',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_2',\n",
       " 'feature_1 + feature_3 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % feature_2 + feature_3 * feature_3 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_1 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_1 + feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_2 + feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_1 + feature_3 % feature_1',\n",
       " 'feature_2 % feature_1 * feature_2 + feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_1 + feature_2 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_2 + feature_3 % feature_2',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_2 + feature_3 * feature_1 + feature_2 % feature_3',\n",
       " 'feature_1 % feature_2 * feature_1 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_2 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_1',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_1 + feature_2 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) * feature_1 + feature_2 % feature_1',\n",
       " 'feature_1 + feature_3 * feature_2 + feature_3 % feature_2',\n",
       " 'feature_1 + feature_2 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % feature_2 + feature_3 * feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_1 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_1 + feature_3 * feature_2 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_3 % feature_2 + feature_3 * feature_2 % feature_2 + feature_3',\n",
       " 'feature_2 + feature_3 % feature_1 * feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) * feature_2 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_1 + feature_3 % feature_3',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * feature_1 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_1 + feature_3 % feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_2 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_1 + feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 * DAYS_SINCE(first_active_month) + feature_1',\n",
       " 'feature_1 + feature_3 % feature_2 * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_3 % feature_2 * feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_3',\n",
       " 'feature_2 + feature_3 % feature_1 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_3 % feature_1 * feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) * feature_1 + feature_3 % feature_2',\n",
       " 'feature_2 % feature_3 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % feature_1 + feature_2 * feature_2',\n",
       " 'DAYS_SINCE(first_active_month) * DAYS_SINCE(first_active_month) % feature_2',\n",
       " 'feature_1 + feature_3 % DAYS_SINCE(first_active_month) * feature_2 % feature_1 + feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_1 + feature_2',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_1 + feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_1 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 % feature_2 + feature_3 * feature_1 + feature_3 % feature_3',\n",
       " 'feature_1 * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_2 % feature_2 + feature_3 * feature_3',\n",
       " 'feature_2 % feature_1 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 % feature_2 * feature_1 + feature_3 % feature_1',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_2 + feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_3 % feature_1',\n",
       " 'feature_1 % feature_2 * feature_2 + feature_3 % feature_2',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_1 + feature_2 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) * DAYS_SINCE(first_active_month) % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 * feature_1 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 * feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_1 + feature_2 % feature_2',\n",
       " 'feature_3 % feature_1 + feature_2 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_1 + feature_3 % feature_2',\n",
       " 'feature_1 + feature_2 % feature_2 + feature_3 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_2 + feature_3 % DAYS_SINCE(first_active_month) * feature_3 % feature_1 + feature_3',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) * feature_2 + feature_3 % feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 * feature_1 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % feature_1 + feature_3 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_2 + feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_2 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_2 + feature_3 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_1 % feature_3',\n",
       " 'feature_1 + feature_3 * feature_1 + feature_3 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'DAYS_SINCE(first_active_month) + feature_3 * feature_1 + feature_3 % feature_1',\n",
       " 'feature_1 + feature_2 % feature_2 * feature_1 + feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_2 % feature_3',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_2 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) % feature_3 * feature_1 + feature_3 % feature_3',\n",
       " 'feature_1 % feature_2 * feature_1 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % feature_2 + feature_3 * feature_2 + feature_3',\n",
       " 'feature_2 + feature_3 % feature_1 + feature_2 * feature_2 + feature_3 % feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_2 * feature_2 % feature_2 + feature_3',\n",
       " 'feature_2 % DAYS_SINCE(first_active_month) * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_2 % feature_1 + feature_3 * feature_3 % feature_2 + feature_3',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_2 % feature_1 + feature_3',\n",
       " 'feature_1 + feature_2 % feature_2 + feature_3 * feature_1 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_1 % feature_3',\n",
       " 'feature_1 + feature_2 % feature_3 * feature_2 % feature_1',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_2 + feature_3 % feature_1',\n",
       " 'feature_1 + feature_3 % feature_1 * feature_3 % feature_1',\n",
       " 'DAYS_SINCE(first_active_month) * DAYS_SINCE(first_active_month) + feature_2',\n",
       " 'feature_2 + feature_3 % feature_2 * feature_3 % feature_2 + feature_3',\n",
       " 'DAYS_SINCE(first_active_month) + feature_1 * DAYS_SINCE(first_active_month) + feature_2',\n",
       " 'feature_1 % feature_2 * feature_1 + feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 + feature_2 * feature_2 % feature_2 + feature_3',\n",
       " 'feature_1 % feature_1 + feature_3 * feature_1 + feature_2 % feature_2',\n",
       " 'DAYS_SINCE(first_active_month) % feature_1 + feature_3 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 % feature_2 + feature_3 * feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 * feature_2 + feature_3 % feature_1 + feature_2',\n",
       " 'feature_1 + feature_3 % feature_2 + feature_3 * feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_2 % feature_3 * feature_2 + feature_3 % feature_3',\n",
       " 'feature_1 + feature_3 % feature_2 + feature_3 * feature_2 % feature_1 + feature_2',\n",
       " 'feature_1 % DAYS_SINCE(first_active_month) * feature_1 + feature_2 % feature_2 + feature_3',\n",
       " 'feature_2 * feature_2 + feature_3 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 % feature_1 + feature_2 * feature_3 % feature_2',\n",
       " 'feature_1 + feature_2 % DAYS_SINCE(first_active_month) * feature_1 + feature_3 % feature_1',\n",
       " 'feature_2 * feature_2 % DAYS_SINCE(first_active_month)',\n",
       " 'feature_1 + feature_3 % feature_3 * feature_3 % feature_1 + feature_2',\n",
       " ...]"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "to_drop"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "hidden": true
   },
   "outputs": [],
   "source": [
    "pruned = lala_feature_matrix.drop(to_drop, axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {
    "hidden": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(123623, 237)"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pruned.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {
    "hidden": true
   },
   "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_1</th>\n",
       "      <th>feature_2</th>\n",
       "      <th>feature_3</th>\n",
       "      <th>DAYS_SINCE(first_active_month)</th>\n",
       "      <th>feature_2 + feature_3</th>\n",
       "      <th>feature_1 + feature_2</th>\n",
       "      <th>feature_1 % feature_2</th>\n",
       "      <th>feature_3 % feature_2</th>\n",
       "      <th>feature_3 % feature_1</th>\n",
       "      <th>feature_2 % feature_1</th>\n",
       "      <th>...</th>\n",
       "      <th>DAYS_SINCE(first_active_month) % feature_1 * feature_1 + feature_2 % feature_2 + feature_3</th>\n",
       "      <th>feature_1 + feature_3 % feature_1 + feature_2 * feature_2 + feature_3 % feature_1</th>\n",
       "      <th>DAYS_SINCE(first_active_month) % feature_1 * feature_2 + feature_3 % feature_1</th>\n",
       "      <th>DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_1 % feature_2 + feature_3</th>\n",
       "      <th>feature_1 + feature_2 % feature_3 * feature_1 + feature_3 % feature_1 + feature_2</th>\n",
       "      <th>DAYS_SINCE(first_active_month) % feature_2 * DAYS_SINCE(first_active_month) % feature_3</th>\n",
       "      <th>feature_1 + feature_2 % feature_1 + feature_3 * feature_1 + feature_3 % feature_1 + feature_2</th>\n",
       "      <th>DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_2 + feature_3 % feature_1</th>\n",
       "      <th>DAYS_SINCE(first_active_month) % feature_1 * feature_1 + feature_2 % feature_3</th>\n",
       "      <th>feature_1 + feature_2 % feature_1 + feature_3 * feature_2 % feature_1 + feature_3</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>card_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>C_ID_0001238066</th>\n",
       "      <td>4.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>522.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C_ID_0001793786</th>\n",
       "      <td>5.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>765.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C_ID_00024e244b</th>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>918.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C_ID_0002709b5a</th>\n",
       "      <td>5.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1284.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>12.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>C_ID_000298032a</th>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>796.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 237 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 feature_1  feature_2  feature_3  \\\n",
       "card_id                                            \n",
       "C_ID_0001238066        4.0        1.0        1.0   \n",
       "C_ID_0001793786        5.0        2.0        2.0   \n",
       "C_ID_00024e244b        2.0        2.0        1.0   \n",
       "C_ID_0002709b5a        5.0        1.0        2.0   \n",
       "C_ID_000298032a        2.0        2.0        1.0   \n",
       "\n",
       "                 DAYS_SINCE(first_active_month)  feature_2 + feature_3  \\\n",
       "card_id                                                                  \n",
       "C_ID_0001238066                           522.0                    2.0   \n",
       "C_ID_0001793786                           765.0                    4.0   \n",
       "C_ID_00024e244b                           918.0                    3.0   \n",
       "C_ID_0002709b5a                          1284.0                    3.0   \n",
       "C_ID_000298032a                           796.0                    3.0   \n",
       "\n",
       "                 feature_1 + feature_2  feature_1 % feature_2  \\\n",
       "card_id                                                         \n",
       "C_ID_0001238066                    5.0                    0.0   \n",
       "C_ID_0001793786                    7.0                    1.0   \n",
       "C_ID_00024e244b                    4.0                    0.0   \n",
       "C_ID_0002709b5a                    6.0                    0.0   \n",
       "C_ID_000298032a                    4.0                    0.0   \n",
       "\n",
       "                 feature_3 % feature_2  feature_3 % feature_1  \\\n",
       "card_id                                                         \n",
       "C_ID_0001238066                    0.0                    1.0   \n",
       "C_ID_0001793786                    0.0                    2.0   \n",
       "C_ID_00024e244b                    1.0                    1.0   \n",
       "C_ID_0002709b5a                    0.0                    2.0   \n",
       "C_ID_000298032a                    1.0                    1.0   \n",
       "\n",
       "                 feature_2 % feature_1  \\\n",
       "card_id                                  \n",
       "C_ID_0001238066                    1.0   \n",
       "C_ID_0001793786                    2.0   \n",
       "C_ID_00024e244b                    0.0   \n",
       "C_ID_0002709b5a                    1.0   \n",
       "C_ID_000298032a                    0.0   \n",
       "\n",
       "                                                       ...                                          \\\n",
       "card_id                                                ...                                           \n",
       "C_ID_0001238066                                        ...                                           \n",
       "C_ID_0001793786                                        ...                                           \n",
       "C_ID_00024e244b                                        ...                                           \n",
       "C_ID_0002709b5a                                        ...                                           \n",
       "C_ID_000298032a                                        ...                                           \n",
       "\n",
       "                 DAYS_SINCE(first_active_month) % feature_1 * feature_1 + feature_2 % feature_2 + feature_3  \\\n",
       "card_id                                                                                                       \n",
       "C_ID_0001238066                                                2.0                                            \n",
       "C_ID_0001793786                                                0.0                                            \n",
       "C_ID_00024e244b                                                0.0                                            \n",
       "C_ID_0002709b5a                                                0.0                                            \n",
       "C_ID_000298032a                                                0.0                                            \n",
       "\n",
       "                 feature_1 + feature_3 % feature_1 + feature_2 * feature_2 + feature_3 % feature_1  \\\n",
       "card_id                                                                                              \n",
       "C_ID_0001238066                                                0.0                                   \n",
       "C_ID_0001793786                                                0.0                                   \n",
       "C_ID_00024e244b                                                3.0                                   \n",
       "C_ID_0002709b5a                                                3.0                                   \n",
       "C_ID_000298032a                                                3.0                                   \n",
       "\n",
       "                 DAYS_SINCE(first_active_month) % feature_1 * feature_2 + feature_3 % feature_1  \\\n",
       "card_id                                                                                           \n",
       "C_ID_0001238066                                                4.0                                \n",
       "C_ID_0001793786                                                0.0                                \n",
       "C_ID_00024e244b                                                0.0                                \n",
       "C_ID_0002709b5a                                               12.0                                \n",
       "C_ID_000298032a                                                0.0                                \n",
       "\n",
       "                 DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_1 % feature_2 + feature_3  \\\n",
       "card_id                                                                                                       \n",
       "C_ID_0001238066                                                0.0                                            \n",
       "C_ID_0001793786                                                1.0                                            \n",
       "C_ID_00024e244b                                                0.0                                            \n",
       "C_ID_0002709b5a                                                0.0                                            \n",
       "C_ID_000298032a                                                2.0                                            \n",
       "\n",
       "                 feature_1 + feature_2 % feature_3 * feature_1 + feature_3 % feature_1 + feature_2  \\\n",
       "card_id                                                                                              \n",
       "C_ID_0001238066                                                0.0                                   \n",
       "C_ID_0001793786                                                0.0                                   \n",
       "C_ID_00024e244b                                                0.0                                   \n",
       "C_ID_0002709b5a                                                0.0                                   \n",
       "C_ID_000298032a                                                0.0                                   \n",
       "\n",
       "                 DAYS_SINCE(first_active_month) % feature_2 * DAYS_SINCE(first_active_month) % feature_3  \\\n",
       "card_id                                                                                                    \n",
       "C_ID_0001238066                                                0.0                                         \n",
       "C_ID_0001793786                                                1.0                                         \n",
       "C_ID_00024e244b                                                0.0                                         \n",
       "C_ID_0002709b5a                                                0.0                                         \n",
       "C_ID_000298032a                                                0.0                                         \n",
       "\n",
       "                 feature_1 + feature_2 % feature_1 + feature_3 * feature_1 + feature_3 % feature_1 + feature_2  \\\n",
       "card_id                                                                                                          \n",
       "C_ID_0001238066                                                0.0                                               \n",
       "C_ID_0001793786                                                0.0                                               \n",
       "C_ID_00024e244b                                                3.0                                               \n",
       "C_ID_0002709b5a                                                6.0                                               \n",
       "C_ID_000298032a                                                3.0                                               \n",
       "\n",
       "                 DAYS_SINCE(first_active_month) % feature_2 + feature_3 * feature_2 + feature_3 % feature_1  \\\n",
       "card_id                                                                                                       \n",
       "C_ID_0001238066                                                0.0                                            \n",
       "C_ID_0001793786                                                4.0                                            \n",
       "C_ID_00024e244b                                                0.0                                            \n",
       "C_ID_0002709b5a                                                0.0                                            \n",
       "C_ID_000298032a                                                1.0                                            \n",
       "\n",
       "                 DAYS_SINCE(first_active_month) % feature_1 * feature_1 + feature_2 % feature_3  \\\n",
       "card_id                                                                                           \n",
       "C_ID_0001238066                                                0.0                                \n",
       "C_ID_0001793786                                                0.0                                \n",
       "C_ID_00024e244b                                                0.0                                \n",
       "C_ID_0002709b5a                                                0.0                                \n",
       "C_ID_000298032a                                                0.0                                \n",
       "\n",
       "                 feature_1 + feature_2 % feature_1 + feature_3 * feature_2 % feature_1 + feature_3  \n",
       "card_id                                                                                             \n",
       "C_ID_0001238066                                                0.0                                  \n",
       "C_ID_0001793786                                                0.0                                  \n",
       "C_ID_00024e244b                                                2.0                                  \n",
       "C_ID_0002709b5a                                                6.0                                  \n",
       "C_ID_000298032a                                                2.0                                  \n",
       "\n",
       "[5 rows x 237 columns]"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pruned.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Merge automated & manual aggs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "add_datepart(temp_train, 'first_active_month', drop=False)\n",
    "add_datepart(temp_test, 'first_active_month', drop=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "new_hist_trans_agg.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def add_additional_aggs3(df):\n",
    "    df['purchase_amount_ptp_new'] = df['MAX(temp_new_trans.purchase_amount)'] - df['MIN(temp_new_trans.purchase_amount)']\n",
    "    df['purchase_amount_ptp_hist'] = df['MAX(temp_hist_trans.purchase_amount)'] - df['MIN(temp_hist_trans.purchase_amount)']\n",
    "    df['card_id_total'] = df['COUNT(temp_hist_trans)'] + df['COUNT(temp_new_trans)']\n",
    "    df['card_id_count_ratio'] = df['COUNT(temp_hist_trans)'] / df['COUNT(temp_new_trans)']\n",
    "    df['purchase_amount_total'] = df['SUM(temp_new_trans.purchase_amount)']+df['SUM(temp_hist_trans.purchase_amount)']\n",
    "    df['purchase_amount_ratio'] = df['SUM(temp_hist_trans.purchase_amount)'] / df['SUM(temp_new_trans.purchase_amount)']\n",
    "    df['purchase_amount_mean_total'] = df['MEAN(temp_new_trans.purchase_amount)']+df['MEAN(temp_hist_trans.purchase_amount)']\n",
    "    # redo with actual min & max\n",
    "    df['purchase_amount_max_total'] = df['MAX(temp_new_trans.purchase_amount)']+df['MAX(temp_hist_trans.purchase_amount)']\n",
    "    df['purchase_amount_min_total'] = df['MIN(temp_new_trans.purchase_amount)']+df['MIN(temp_hist_trans.purchase_amount)']\n",
    "    df['avg_spend_per_transaction'] = df['purchase_amount_total']/df['card_id_total']\n",
    "    df['month_diff_mean_total'] = df['MEAN(temp_new_trans.month_diff)']+df['MEAN(temp_hist_trans.month_diff)']\n",
    "    df['month_diff_mean_ratio'] = df['MEAN(temp_hist_trans.month_diff)'] / df['MEAN(temp_new_trans.month_diff)']\n",
    "\n",
    "    df['month_lag_ratio'] = df['SUM(temp_hist_trans.month_lag)'] / df['SUM(temp_new_trans.month_lag)']\n",
    "    df['month_lag_sum'] = df['SUM(temp_hist_trans.month_lag)'] + df['SUM(temp_new_trans.month_lag)']\n",
    "    df['month_lag_mean_total'] = df['MEAN(temp_new_trans.month_lag)']+df['MEAN(temp_hist_trans.month_lag)']\n",
    "    # redo with actual min & max\n",
    "    df['month_lag_max_total'] = df['MAX(temp_new_trans.month_lag)']+df['MAX(temp_hist_trans.month_lag)']\n",
    "    df['month_lag_min_total'] = df['MIN(temp_new_trans.month_lag)']+df['MIN(temp_hist_trans.month_lag)']\n",
    "    \n",
    "    df['installments_ratio'] = df['SUM(temp_hist_trans.installments)'] / df['SUM(temp_new_trans.installments)']\n",
    "    df['installments_sum'] = df['SUM(temp_hist_trans.installments)'] + df['SUM(temp_new_trans.installments)']\n",
    "    df['installments_mean_total'] = df['MEAN(temp_new_trans.installments)']+df['MEAN(temp_hist_trans.installments)']\n",
    "    # redo with actual min & max\n",
    "    df['installments_max_total'] = df['MAX(temp_new_trans.installments)']+df['MAX(temp_hist_trans.installments)']\n",
    "    df['installments_min_total'] = df['MIN(temp_new_trans.installments)']+df['MIN(temp_hist_trans.installments)']\n",
    "    \n",
    "    df['duration_sum_ratio'] = df['SUM(temp_hist_trans.duration)'] / df['SUM(temp_new_trans.duration)']\n",
    "    df['duration_mean_total'] = df['MEAN(temp_new_trans.duration)']+df['MEAN(temp_hist_trans.duration)']\n",
    "    # redo with actual min & max\n",
    "    df['duration_max_total'] = df['MAX(temp_new_trans.duration)']+df['MAX(temp_hist_trans.duration)']\n",
    "    df['duration_min_total'] = df['MIN(temp_new_trans.duration)']+df['MIN(temp_hist_trans.duration)']\n",
    "    \n",
    "    df['amount_month_ratio_sum_ratio'] = df['SUM(temp_hist_trans.amount_month_ratio)'] / df['SUM(temp_new_trans.amount_month_ratio)']\n",
    "    df['amount_month_ratio_mean_total'] = df['MEAN(temp_new_trans.amount_month_ratio)']+df['MEAN(temp_hist_trans.amount_month_ratio)']\n",
    "    # redo with actual min & max\n",
    "    df['amount_month_ratio_max_total'] = df['MAX(temp_new_trans.amount_month_ratio)']+df['MAX(temp_hist_trans.amount_month_ratio)']\n",
    "    df['amount_month_ratio_min_total'] = df['MIN(temp_new_trans.amount_month_ratio)']+df['MIN(temp_hist_trans.amount_month_ratio)']\n",
    "    \n",
    "    df['category_1_mean'] = df['category_1_mean']+df['category_1_mean_old']\n",
    "    df['price_total'] = df['purchase_amount_total'] / df['installments_sum']\n",
    "    df['price_mean'] = df['purchase_amount_mean_total'] / df['installments_mean_total']\n",
    "    df['price_max'] = df['purchase_amount_max_total'] / df['installments_max_total']\n",
    "\n",
    "    df['CLV_new'] = df['COUNT(temp_new_trans)'] * df['SUM(temp_new_trans.purchase_amount)'] / df['MEAN(temp_new_trans.month_diff)']\n",
    "    df['CLV_hist'] = df['COUNT(temp_hist_trans)'] * df['SUM(temp_hist_trans.purchase_amount)'] / df['MEAN(temp_hist_trans.month_diff)']\n",
    "    \n",
    "    df['CLV_ratio'] = df['CLV_hist'] / df['CLV_new']\n",
    "    df['elapsed_time'] = (datetime.datetime.today() - df['first_active_month']).dt.days\n",
    "\n",
    "    df['hist_first_buy'] = (df['purchase_date_min_old'] - df['first_active_month']).dt.days\n",
    "    df['new_first_buy'] = (df['purchase_date_min'] - df['first_active_month']).dt.days\n",
    "    df['hist_last_buy'] = (df['purchase_date_max_old'] - df['first_active_month']).dt.days\n",
    "    df['new_last_buy'] = (df['purchase_date_max'] - df['first_active_month']).dt.days\n",
    "    df['purchased_before_issue'] = df['hist_first_buy'] < 0\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def join_dfs(left, right, left_on, right_on=None, suffix='_old'):\n",
    "    if right_on is None: right_on = left_on\n",
    "    return left.merge(right, how='left', left_on=left_on, right_on=right_on, suffixes=(\"\", suffix))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df = join_dfs(temp_train, new_hist_trans_agg, left_on='card_id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df = join_dfs(train_df, hist_trans_agg, left_on='card_id', suffix='_old')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df = join_dfs(train_df, feature_matrix, left_on='card_id')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "def add_days_feature_interaction(df):\n",
    "    # to datetime\n",
    "    df['first_active_month'] = pd.to_datetime(df['first_active_month'])\n",
    "\n",
    "    # datetime features\n",
    "    df['quarter'] = df['first_active_month'].dt.quarter\n",
    "    df['elapsed_time'] = (datetime.datetime.today() - df['first_active_month']).dt.days\n",
    "\n",
    "    df['days_feature1'] = df['elapsed_time'] * df['feature_1']\n",
    "    df['days_feature2'] = df['elapsed_time'] * df['feature_2']\n",
    "    df['days_feature3'] = df['elapsed_time'] * df['feature_3']\n",
    "\n",
    "    df['days_feature1_ratio'] = df['feature_1'] / df['elapsed_time']\n",
    "    df['days_feature2_ratio'] = df['feature_2'] / df['elapsed_time']\n",
    "    df['days_feature3_ratio'] = df['feature_3'] / df['elapsed_time']\n",
    "    df['feature_sum'] = df['feature_1'] + df['feature_2'] + df['feature_3']\n",
    "    df['feature_mean'] = df['feature_sum']/3\n",
    "    df['feature_max'] = df[['feature_1', 'feature_2', 'feature_3']].max(axis=1)\n",
    "    df['feature_min'] = df[['feature_1', 'feature_2', 'feature_3']].min(axis=1)\n",
    "    df['feature_var'] = df[['feature_1', 'feature_2', 'feature_3']].std(axis=1)\n",
    "    return df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df = add_additional_aggs3(train_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df = add_days_feature_interaction(train_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_df = join_dfs(temp_test, new_hist_trans_agg, left_on='card_id')\n",
    "test_df = join_dfs(test_df, hist_trans_agg, left_on='card_id', suffix='_old')\n",
    "test_df = join_dfs(test_df, feature_matrix, left_on='card_id')\n",
    "test_df = add_additional_aggs3(test_df)\n",
    "test_df = add_days_feature_interaction(test_df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df['outliers'] = 0\n",
    "train_df.loc[train_df['target'] < -30, 'outliers'] = 1\n",
    "train_df['outliers'].value_counts()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for f in ['feature_1','feature_2','feature_3']:\n",
    "    order_label = train_df.groupby([f])['outliers'].mean()\n",
    "    train_df[f] = train_df[f].map(order_label)\n",
    "    test_df[f] = test_df[f].map(order_label)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "train_df.to_feather('train_ft')\n",
    "test_df.to_feather('test_ft')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.6.6"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": true,
   "toc_position": {},
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}