Customer Subscription Behaviour Analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Business problem & dataset overview"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We are working for a fintech company that wants to provide customers with a paid mobile app subscription that will allow them to track their finances.\n",
    "\n",
    "To attract customers, the company releases a free version of the app.\n",
    "\n",
    "The task is to identify which users will most likely NOT purchase the paid version of the app, so that additional offers can be provided to them. Accuracy is key, because increased targeted advertising is costly.\n",
    "\n",
    "Our dataset features:\n",
    "- User ID\n",
    "- Day and time at which the user first opened the app\n",
    "- Day of the week (numerical) at which the user first opened the app\n",
    "- Hour of the day at which the user first opened the app\n",
    "- User age\n",
    "- Screen list: every screen name visited by user in the app\n",
    "- Number of screens visited in the app by the user\n",
    "- Whether the user played the available minigame\n",
    "- Whether the user gave a \"like\" rating to the app\n",
    "- Whether the user used the app's premium feature\n",
    "- Whether the user enrolled in the paid product\n",
    "- The date of enrolment (if any)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 143,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "from dateutil import parser"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 144,
   "metadata": {
    "scrolled": 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>user</th>\n",
       "      <th>first_open</th>\n",
       "      <th>dayofweek</th>\n",
       "      <th>hour</th>\n",
       "      <th>age</th>\n",
       "      <th>screen_list</th>\n",
       "      <th>numscreens</th>\n",
       "      <th>minigame</th>\n",
       "      <th>used_premium_feature</th>\n",
       "      <th>enrolled</th>\n",
       "      <th>enrolled_date</th>\n",
       "      <th>liked</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>235136</td>\n",
       "      <td>2012-12-27 02:14:51.273</td>\n",
       "      <td>3</td>\n",
       "      <td>02:00:00</td>\n",
       "      <td>23</td>\n",
       "      <td>idscreen,joinscreen,Cycle,product_review,ScanP...</td>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>333588</td>\n",
       "      <td>2012-12-02 01:16:00.905</td>\n",
       "      <td>6</td>\n",
       "      <td>01:00:00</td>\n",
       "      <td>24</td>\n",
       "      <td>joinscreen,product_review,product_review2,Scan...</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>254414</td>\n",
       "      <td>2013-03-19 19:19:09.157</td>\n",
       "      <td>1</td>\n",
       "      <td>19:00:00</td>\n",
       "      <td>23</td>\n",
       "      <td>Splash,Cycle,Loan</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>234192</td>\n",
       "      <td>2013-07-05 16:08:46.354</td>\n",
       "      <td>4</td>\n",
       "      <td>16:00:00</td>\n",
       "      <td>28</td>\n",
       "      <td>product_review,Home,product_review,Loan3,Finan...</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2013-07-05 16:11:49.513</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>51549</td>\n",
       "      <td>2013-02-26 18:50:48.661</td>\n",
       "      <td>1</td>\n",
       "      <td>18:00:00</td>\n",
       "      <td>31</td>\n",
       "      <td>idscreen,joinscreen,Cycle,Credit3Container,Sca...</td>\n",
       "      <td>32</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2013-02-26 18:56:37.841</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>56480</td>\n",
       "      <td>2013-04-03 09:58:15.752</td>\n",
       "      <td>2</td>\n",
       "      <td>09:00:00</td>\n",
       "      <td>20</td>\n",
       "      <td>idscreen,Cycle,Home,ScanPreview,VerifyPhone,Ve...</td>\n",
       "      <td>14</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2013-04-03 09:59:03.291</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>144649</td>\n",
       "      <td>2012-12-25 02:33:18.461</td>\n",
       "      <td>1</td>\n",
       "      <td>02:00:00</td>\n",
       "      <td>35</td>\n",
       "      <td>product_review,product_review2,ScanPreview</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>249366</td>\n",
       "      <td>2012-12-11 03:07:49.875</td>\n",
       "      <td>1</td>\n",
       "      <td>03:00:00</td>\n",
       "      <td>26</td>\n",
       "      <td>Splash,Cycle,Home,Credit3Container,Credit3Dash...</td>\n",
       "      <td>41</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>372004</td>\n",
       "      <td>2013-03-20 14:22:01.569</td>\n",
       "      <td>2</td>\n",
       "      <td>14:00:00</td>\n",
       "      <td>29</td>\n",
       "      <td>product_review,product_review2,ScanPreview,Ver...</td>\n",
       "      <td>33</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2013-04-27 22:24:54.542</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>338013</td>\n",
       "      <td>2013-04-26 18:22:16.013</td>\n",
       "      <td>4</td>\n",
       "      <td>18:00:00</td>\n",
       "      <td>26</td>\n",
       "      <td>Home,Loan2,product_review,product_review,produ...</td>\n",
       "      <td>19</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>2013-04-26 18:31:58.923</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     user               first_open  dayofweek       hour  age  \\\n",
       "0  235136  2012-12-27 02:14:51.273          3   02:00:00   23   \n",
       "1  333588  2012-12-02 01:16:00.905          6   01:00:00   24   \n",
       "2  254414  2013-03-19 19:19:09.157          1   19:00:00   23   \n",
       "3  234192  2013-07-05 16:08:46.354          4   16:00:00   28   \n",
       "4   51549  2013-02-26 18:50:48.661          1   18:00:00   31   \n",
       "5   56480  2013-04-03 09:58:15.752          2   09:00:00   20   \n",
       "6  144649  2012-12-25 02:33:18.461          1   02:00:00   35   \n",
       "7  249366  2012-12-11 03:07:49.875          1   03:00:00   26   \n",
       "8  372004  2013-03-20 14:22:01.569          2   14:00:00   29   \n",
       "9  338013  2013-04-26 18:22:16.013          4   18:00:00   26   \n",
       "\n",
       "                                         screen_list  numscreens  minigame  \\\n",
       "0  idscreen,joinscreen,Cycle,product_review,ScanP...          15         0   \n",
       "1  joinscreen,product_review,product_review2,Scan...          13         0   \n",
       "2                                  Splash,Cycle,Loan           3         0   \n",
       "3  product_review,Home,product_review,Loan3,Finan...          40         0   \n",
       "4  idscreen,joinscreen,Cycle,Credit3Container,Sca...          32         0   \n",
       "5  idscreen,Cycle,Home,ScanPreview,VerifyPhone,Ve...          14         0   \n",
       "6         product_review,product_review2,ScanPreview           3         0   \n",
       "7  Splash,Cycle,Home,Credit3Container,Credit3Dash...          41         0   \n",
       "8  product_review,product_review2,ScanPreview,Ver...          33         1   \n",
       "9  Home,Loan2,product_review,product_review,produ...          19         0   \n",
       "\n",
       "   used_premium_feature  enrolled            enrolled_date  liked  \n",
       "0                     0         0                      NaN      0  \n",
       "1                     0         0                      NaN      0  \n",
       "2                     1         0                      NaN      1  \n",
       "3                     0         1  2013-07-05 16:11:49.513      0  \n",
       "4                     0         1  2013-02-26 18:56:37.841      1  \n",
       "5                     0         1  2013-04-03 09:59:03.291      0  \n",
       "6                     0         0                      NaN      0  \n",
       "7                     1         0                      NaN      0  \n",
       "8                     1         1  2013-04-27 22:24:54.542      0  \n",
       "9                     0         1  2013-04-26 18:31:58.923      0  "
      ]
     },
     "execution_count": 144,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset = pd.read_csv(\"P39-CS3-Data/appdata10.csv\")\n",
    "dataset.head(n=10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 145,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 50000 entries, 0 to 49999\n",
      "Data columns (total 12 columns):\n",
      " #   Column                Non-Null Count  Dtype \n",
      "---  ------                --------------  ----- \n",
      " 0   user                  50000 non-null  int64 \n",
      " 1   first_open            50000 non-null  object\n",
      " 2   dayofweek             50000 non-null  int64 \n",
      " 3   hour                  50000 non-null  object\n",
      " 4   age                   50000 non-null  int64 \n",
      " 5   screen_list           50000 non-null  object\n",
      " 6   numscreens            50000 non-null  int64 \n",
      " 7   minigame              50000 non-null  int64 \n",
      " 8   used_premium_feature  50000 non-null  int64 \n",
      " 9   enrolled              50000 non-null  int64 \n",
      " 10  enrolled_date         31074 non-null  object\n",
      " 11  liked                 50000 non-null  int64 \n",
      "dtypes: int64(8), object(4)\n",
      "memory usage: 4.6+ MB\n"
     ]
    }
   ],
   "source": [
    "dataset.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 146,
   "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>user</th>\n",
       "      <th>dayofweek</th>\n",
       "      <th>age</th>\n",
       "      <th>numscreens</th>\n",
       "      <th>minigame</th>\n",
       "      <th>used_premium_feature</th>\n",
       "      <th>enrolled</th>\n",
       "      <th>liked</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.00000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>186889.729900</td>\n",
       "      <td>3.029860</td>\n",
       "      <td>31.72436</td>\n",
       "      <td>21.095900</td>\n",
       "      <td>0.107820</td>\n",
       "      <td>0.172020</td>\n",
       "      <td>0.621480</td>\n",
       "      <td>0.165000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>107768.520361</td>\n",
       "      <td>2.031997</td>\n",
       "      <td>10.80331</td>\n",
       "      <td>15.728812</td>\n",
       "      <td>0.310156</td>\n",
       "      <td>0.377402</td>\n",
       "      <td>0.485023</td>\n",
       "      <td>0.371184</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>13.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>16.00000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>93526.750000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>24.00000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>187193.500000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>29.00000</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>279984.250000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>37.00000</td>\n",
       "      <td>28.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>373662.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>101.00000</td>\n",
       "      <td>325.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                user     dayofweek          age    numscreens      minigame  \\\n",
       "count   50000.000000  50000.000000  50000.00000  50000.000000  50000.000000   \n",
       "mean   186889.729900      3.029860     31.72436     21.095900      0.107820   \n",
       "std    107768.520361      2.031997     10.80331     15.728812      0.310156   \n",
       "min        13.000000      0.000000     16.00000      1.000000      0.000000   \n",
       "25%     93526.750000      1.000000     24.00000     10.000000      0.000000   \n",
       "50%    187193.500000      3.000000     29.00000     18.000000      0.000000   \n",
       "75%    279984.250000      5.000000     37.00000     28.000000      0.000000   \n",
       "max    373662.000000      6.000000    101.00000    325.000000      1.000000   \n",
       "\n",
       "       used_premium_feature      enrolled         liked  \n",
       "count          50000.000000  50000.000000  50000.000000  \n",
       "mean               0.172020      0.621480      0.165000  \n",
       "std                0.377402      0.485023      0.371184  \n",
       "min                0.000000      0.000000      0.000000  \n",
       "25%                0.000000      0.000000      0.000000  \n",
       "50%                0.000000      1.000000      0.000000  \n",
       "75%                0.000000      1.000000      0.000000  \n",
       "max                1.000000      1.000000      1.000000  "
      ]
     },
     "execution_count": 146,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.describe() # generate descriptive statistics"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data pre-processing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 147,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Formatting the \"hour\" column to obtain descriptive statistics\n",
    "dataset[\"hour\"] = dataset.hour.str.slice(1, 3).astype(int) # 1st and 2nd numbers"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 148,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset2 = dataset.copy().drop(columns = [\"user\", \"screen_list\", \"enrolled_date\", \"first_open\", \"enrolled\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 149,
   "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>dayofweek</th>\n",
       "      <th>hour</th>\n",
       "      <th>age</th>\n",
       "      <th>numscreens</th>\n",
       "      <th>minigame</th>\n",
       "      <th>used_premium_feature</th>\n",
       "      <th>liked</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>23</td>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>24</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1</td>\n",
       "      <td>19</td>\n",
       "      <td>23</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4</td>\n",
       "      <td>16</td>\n",
       "      <td>28</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>1</td>\n",
       "      <td>18</td>\n",
       "      <td>31</td>\n",
       "      <td>32</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49995</th>\n",
       "      <td>3</td>\n",
       "      <td>13</td>\n",
       "      <td>32</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49996</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>35</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49997</th>\n",
       "      <td>2</td>\n",
       "      <td>22</td>\n",
       "      <td>39</td>\n",
       "      <td>25</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49998</th>\n",
       "      <td>6</td>\n",
       "      <td>12</td>\n",
       "      <td>27</td>\n",
       "      <td>26</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>49999</th>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>25</td>\n",
       "      <td>26</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>50000 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       dayofweek  hour  age  numscreens  minigame  used_premium_feature  liked\n",
       "0              3     2   23          15         0                     0      0\n",
       "1              6     1   24          13         0                     0      0\n",
       "2              1    19   23           3         0                     1      1\n",
       "3              4    16   28          40         0                     0      0\n",
       "4              1    18   31          32         0                     0      1\n",
       "...          ...   ...  ...         ...       ...                   ...    ...\n",
       "49995          3    13   32          13         0                     0      0\n",
       "49996          1     0   35           4         0                     1      0\n",
       "49997          2    22   39          25         0                     0      0\n",
       "49998          6    12   27          26         0                     0      0\n",
       "49999          4     1   25          26         0                     0      1\n",
       "\n",
       "[50000 rows x 7 columns]"
      ]
     },
     "execution_count": 149,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset2 # only contains numerical columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 150,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 7 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Generating histograms to visualize the dataset\n",
    "subplot_title_info = {'fontsize': 15,\n",
    "        'fontweight' : 5,\n",
    "        'verticalalignment': 'baseline',\n",
    "        'horizontalalignment': 'center'}\n",
    "hour_ticks = [0, 3, 6, 9, 12, 15, 18, 21]\n",
    "\n",
    "plt.suptitle(\"Histogram of Numerical Columns\", fontsize = 20,\n",
    "            x = 1, y = 3)\n",
    "for i in range(1, dataset2.shape[1] + 1):\n",
    "    plt.subplot(3, 3, i) # dimensions of our plot\n",
    "    f = plt.gca() # get current axis index\n",
    "    f.set_title(dataset2.columns.values[i - 1], subplot_title_info)\n",
    "    f.axes.get_yaxis().set_visible(False)\n",
    "    vals = np.size(dataset2.iloc[:, i - 1].unique())\n",
    "    plt.hist(dataset2.iloc[:, i-1], bins = vals, color = \"#3F5D7D\")\n",
    "    if (i == 2):\n",
    "        f.axes.set_xticks(ticks = hour_ticks)\n",
    "        \n",
    "plt.tight_layout(rect = [0, 1, 2, 3])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 151,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = pd.read_csv(\"P39-CS3-Data/appdata10.csv\")\n",
    "dataset[\"hour\"] = dataset.hour.str.slice(1, 3).astype(str) # 1st and 2nd numbers\n",
    "dataset2 = dataset.copy().drop(columns = [\"user\", \"screen_list\", \"enrolled_date\", \"first_open\", \"enrolled\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 152,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f88a8ee3210>"
      ]
     },
     "execution_count": 152,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x720 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "dataset2.corrwith(dataset.enrolled).plot.bar(figsize = (20, 10),\n",
    "                                            fontsize = 15,\n",
    "                                            rot = 45,\n",
    "                                            grid = True,\n",
    "                                            title = \"Correlation with response variable\")\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 153,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x7f88a90f6810>"
      ]
     },
     "execution_count": 153,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x1080 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Generating a correlation matrix\n",
    "sns.set(style = \"white\", font_scale = 2)\n",
    "corr = dataset2.corr()\n",
    "\n",
    "# generate mask for upper triangle\n",
    "# this is because the matrix is symmetric\n",
    "mask = np.zeros_like(corr, dtype = np.bool)\n",
    "mask[np.triu_indices_from(mask)] = True\n",
    "\n",
    "f, ax = plt.subplots(figsize = (18, 15))\n",
    "f.suptitle(\"Correlation Matrix\", fontsize = 40)\n",
    "\n",
    "# Generating a customized diverging color map\n",
    "cmap = sns.diverging_palette(220, 10, as_cmap = True)\n",
    "\n",
    "# Draw heatmap with the mask and correct aspect ratio\n",
    "sns.heatmap(corr, mask = mask, cmap = cmap, vmax = 0.3, center = 0,\n",
    "          square = True, linewidths = 0.5, cbar_kws = {\"shrink\":.5})"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Feature Engineering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 154,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "user                     int64\n",
       "first_open              object\n",
       "dayofweek                int64\n",
       "hour                    object\n",
       "age                      int64\n",
       "screen_list             object\n",
       "numscreens               int64\n",
       "minigame                 int64\n",
       "used_premium_feature     int64\n",
       "enrolled                 int64\n",
       "enrolled_date           object\n",
       "liked                    int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 154,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 155,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Convert \"first_open\" into datetime object\n",
    "dataset[\"first_open\"] = [parser.parse(row_date) for row_date in dataset[\"first_open\"]]\n",
    "dataset[\"enrolled_date\"] = [parser.parse(row_date) if isinstance(row_date, str) else row_date for row_date in dataset[\"enrolled_date\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 156,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "user                             int64\n",
       "first_open              datetime64[ns]\n",
       "dayofweek                        int64\n",
       "hour                            object\n",
       "age                              int64\n",
       "screen_list                     object\n",
       "numscreens                       int64\n",
       "minigame                         int64\n",
       "used_premium_feature             int64\n",
       "enrolled                         int64\n",
       "enrolled_date           datetime64[ns]\n",
       "liked                            int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 156,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 157,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Add a new column for time between enrolment and first open of the app\n",
    "dataset[\"difference\"] = (dataset.enrolled_date - dataset.first_open).astype(\"timedelta64[h]\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 158,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(dataset[\"difference\"].dropna(),  color = \"#3F5D7D\")\n",
    "plt.title(\"Distribution of Time Since Enrolment\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 159,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.hist(dataset[\"difference\"].dropna(),  color = \"#3F5D7D\",\n",
    "        range = [0, 100])\n",
    "plt.title(\"Distribution of Time Since Enrolment First 100 Hours\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 160,
   "metadata": {},
   "outputs": [],
   "source": [
    "# For those people outside the 48 hour cutoff mark,\n",
    "# we will mark them as unenrolled\n",
    "dataset.loc[dataset.difference > 48, \"enrolled\"] = 0"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 161,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = dataset.drop(columns = [\"difference\", \"enrolled_date\", \"first_open\"])\n",
    "# drop unused columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 162,
   "metadata": {},
   "outputs": [],
   "source": [
    "top_screens = pd.read_csv(\"P39-CS3-Data/top_screens.csv\").top_screens.values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 163,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create as many commas as there are screens\n",
    "# this will help us count the number of different screens in the dataset\n",
    "dataset[\"screen_list\"] = dataset.screen_list.astype(str) + \",\"\n",
    "\n",
    "for sc in top_screens:\n",
    "    dataset[sc] = dataset.screen_list.str.contains(sc).astype(int) # 0 or 1 values\n",
    "    dataset[\"screen_list\"] = dataset.screen_list.str.replace(sc+\",\", \"\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 164,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset[\"Other\"] = dataset.screen_list.str.count(\",\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 165,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = dataset.drop(columns = [\"screen_list\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 166,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Funnels = group of screens that belong to the same set\n",
    "# we do not want correlated screens\n",
    "# we will group screens by funnel\n",
    "\n",
    "savings_screens = [\"Saving1\",\n",
    "                  \"Saving2\",\n",
    "                  \"Saving2Amount\",\n",
    "                  \"Saving4\",\n",
    "                  \"Saving5\",\n",
    "                  \"Saving6\",\n",
    "                  \"Saving7\",\n",
    "                  \"Saving8\",\n",
    "                  \"Saving9\",\n",
    "                  \"Saving10\"]\n",
    "\n",
    "dataset[\"SavingCount\"] = dataset[savings_screens].sum(axis = 1)\n",
    "dataset = dataset.drop(columns = savings_screens)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 167,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['user', 'dayofweek', 'hour', 'age', 'numscreens', 'minigame',\n",
       "       'used_premium_feature', 'enrolled', 'liked', 'Loan2', 'location',\n",
       "       'Institutions', 'Credit3Container', 'VerifyPhone', 'BankVerification',\n",
       "       'VerifyDateOfBirth', 'ProfilePage', 'VerifyCountry', 'Cycle',\n",
       "       'idscreen', 'Credit3Dashboard', 'Loan3', 'CC1Category', 'Splash',\n",
       "       'Loan', 'CC1', 'RewardsContainer', 'Credit3', 'Credit1', 'EditProfile',\n",
       "       'Credit2', 'Finances', 'CC3', 'Alerts', 'Leaderboard', 'VerifyMobile',\n",
       "       'VerifyHousing', 'RewardDetail', 'VerifyHousingAmount',\n",
       "       'ProfileMaritalStatus', 'ProfileChildren ', 'ProfileEducation',\n",
       "       'ProfileEducationMajor', 'Rewards', 'AccountView', 'VerifyAnnualIncome',\n",
       "       'VerifyIncomeType', 'ProfileJobTitle', 'Login',\n",
       "       'ProfileEmploymentLength', 'WebView', 'SecurityModal', 'Loan4',\n",
       "       'ResendToken', 'TransactionList', 'NetworkFailure', 'ListPicker',\n",
       "       'Other', 'SavingCount'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 167,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 168,
   "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>user</th>\n",
       "      <th>dayofweek</th>\n",
       "      <th>hour</th>\n",
       "      <th>age</th>\n",
       "      <th>numscreens</th>\n",
       "      <th>minigame</th>\n",
       "      <th>used_premium_feature</th>\n",
       "      <th>enrolled</th>\n",
       "      <th>liked</th>\n",
       "      <th>location</th>\n",
       "      <th>...</th>\n",
       "      <th>SecurityModal</th>\n",
       "      <th>ResendToken</th>\n",
       "      <th>TransactionList</th>\n",
       "      <th>NetworkFailure</th>\n",
       "      <th>ListPicker</th>\n",
       "      <th>Other</th>\n",
       "      <th>SavingCount</th>\n",
       "      <th>CMCount</th>\n",
       "      <th>CCCount</th>\n",
       "      <th>LoansCount</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>235136</td>\n",
       "      <td>3</td>\n",
       "      <td>02</td>\n",
       "      <td>23</td>\n",
       "      <td>15</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>333588</td>\n",
       "      <td>6</td>\n",
       "      <td>01</td>\n",
       "      <td>24</td>\n",
       "      <td>13</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>254414</td>\n",
       "      <td>1</td>\n",
       "      <td>19</td>\n",
       "      <td>23</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>234192</td>\n",
       "      <td>4</td>\n",
       "      <td>16</td>\n",
       "      <td>28</td>\n",
       "      <td>40</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>51549</td>\n",
       "      <td>1</td>\n",
       "      <td>18</td>\n",
       "      <td>31</td>\n",
       "      <td>32</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>10</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>56480</td>\n",
       "      <td>2</td>\n",
       "      <td>09</td>\n",
       "      <td>20</td>\n",
       "      <td>14</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>144649</td>\n",
       "      <td>1</td>\n",
       "      <td>02</td>\n",
       "      <td>35</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>7 rows × 50 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     user  dayofweek hour  age  numscreens  minigame  used_premium_feature  \\\n",
       "0  235136          3   02   23          15         0                     0   \n",
       "1  333588          6   01   24          13         0                     0   \n",
       "2  254414          1   19   23           3         0                     1   \n",
       "3  234192          4   16   28          40         0                     0   \n",
       "4   51549          1   18   31          32         0                     0   \n",
       "5   56480          2   09   20          14         0                     0   \n",
       "6  144649          1   02   35           3         0                     0   \n",
       "\n",
       "   enrolled  liked  location  ...  SecurityModal  ResendToken  \\\n",
       "0         0      0         0  ...              0            0   \n",
       "1         0      0         1  ...              0            0   \n",
       "2         0      1         0  ...              0            0   \n",
       "3         1      0         1  ...              0            0   \n",
       "4         1      1         0  ...              0            0   \n",
       "5         1      0         0  ...              0            0   \n",
       "6         0      0         0  ...              0            0   \n",
       "\n",
       "   TransactionList  NetworkFailure  ListPicker  Other  SavingCount  CMCount  \\\n",
       "0                0               0           0      7            0        0   \n",
       "1                0               0           0      5            0        0   \n",
       "2                0               0           0      0            0        0   \n",
       "3                0               0           0      6            0        3   \n",
       "4                0               0           0     10            0        2   \n",
       "5                0               0           0      6            0        2   \n",
       "6                0               0           0      3            0        0   \n",
       "\n",
       "   CCCount  LoansCount  \n",
       "0        0           1  \n",
       "1        0           1  \n",
       "2        0           1  \n",
       "3        0           1  \n",
       "4        0           1  \n",
       "5        0           1  \n",
       "6        0           0  \n",
       "\n",
       "[7 rows x 50 columns]"
      ]
     },
     "execution_count": 168,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cm_screens = [\"Credit1\",\n",
    "             \"Credit2\",\n",
    "             \"Credit3\",\n",
    "             \"Credit3Container\",\n",
    "             \"Credit3Dashboard\"]\n",
    "cc_screens = [\"CC1\",\n",
    "             \"CC1Category\",\n",
    "             \"CC3\"]\n",
    "loan_screens = [\"Loan\",\n",
    "               \"Loan2\",\n",
    "               \"Loan3\",\n",
    "               \"Loan4\"]\n",
    "dataset[\"CMCount\"] = dataset[cm_screens].sum(axis = 1)\n",
    "dataset[\"CCCount\"] = dataset[cc_screens].sum(axis = 1)\n",
    "dataset[\"LoansCount\"] = dataset[loan_screens].sum(axis = 1)\n",
    "dataset = dataset.drop(columns = cm_screens)\n",
    "dataset = dataset.drop(columns = cc_screens)\n",
    "dataset = dataset.drop(columns = loan_screens)\n",
    "dataset.head(n=7)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 169,
   "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>user</th>\n",
       "      <th>dayofweek</th>\n",
       "      <th>age</th>\n",
       "      <th>numscreens</th>\n",
       "      <th>minigame</th>\n",
       "      <th>used_premium_feature</th>\n",
       "      <th>enrolled</th>\n",
       "      <th>liked</th>\n",
       "      <th>location</th>\n",
       "      <th>Institutions</th>\n",
       "      <th>...</th>\n",
       "      <th>SecurityModal</th>\n",
       "      <th>ResendToken</th>\n",
       "      <th>TransactionList</th>\n",
       "      <th>NetworkFailure</th>\n",
       "      <th>ListPicker</th>\n",
       "      <th>Other</th>\n",
       "      <th>SavingCount</th>\n",
       "      <th>CMCount</th>\n",
       "      <th>CCCount</th>\n",
       "      <th>LoansCount</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.00000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.00000</td>\n",
       "      <td>50000.000000</td>\n",
       "      <td>50000.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>186889.729900</td>\n",
       "      <td>3.029860</td>\n",
       "      <td>31.72436</td>\n",
       "      <td>21.095900</td>\n",
       "      <td>0.107820</td>\n",
       "      <td>0.172020</td>\n",
       "      <td>0.497000</td>\n",
       "      <td>0.165000</td>\n",
       "      <td>0.517760</td>\n",
       "      <td>0.293660</td>\n",
       "      <td>...</td>\n",
       "      <td>0.014220</td>\n",
       "      <td>0.013340</td>\n",
       "      <td>0.013400</td>\n",
       "      <td>0.008200</td>\n",
       "      <td>0.007580</td>\n",
       "      <td>6.214260</td>\n",
       "      <td>0.365020</td>\n",
       "      <td>0.92776</td>\n",
       "      <td>0.176860</td>\n",
       "      <td>0.788400</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>107768.520361</td>\n",
       "      <td>2.031997</td>\n",
       "      <td>10.80331</td>\n",
       "      <td>15.728812</td>\n",
       "      <td>0.310156</td>\n",
       "      <td>0.377402</td>\n",
       "      <td>0.499996</td>\n",
       "      <td>0.371184</td>\n",
       "      <td>0.499689</td>\n",
       "      <td>0.455443</td>\n",
       "      <td>...</td>\n",
       "      <td>0.118398</td>\n",
       "      <td>0.114727</td>\n",
       "      <td>0.114981</td>\n",
       "      <td>0.090183</td>\n",
       "      <td>0.086733</td>\n",
       "      <td>3.672561</td>\n",
       "      <td>1.405511</td>\n",
       "      <td>1.21751</td>\n",
       "      <td>0.612787</td>\n",
       "      <td>0.677462</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>13.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>16.00000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>93526.750000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>24.00000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>187193.500000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>29.00000</td>\n",
       "      <td>18.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>279984.250000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>37.00000</td>\n",
       "      <td>28.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.00000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>373662.000000</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>101.00000</td>\n",
       "      <td>325.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>...</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>35.000000</td>\n",
       "      <td>10.000000</td>\n",
       "      <td>5.00000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>3.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>8 rows × 49 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                user     dayofweek          age    numscreens      minigame  \\\n",
       "count   50000.000000  50000.000000  50000.00000  50000.000000  50000.000000   \n",
       "mean   186889.729900      3.029860     31.72436     21.095900      0.107820   \n",
       "std    107768.520361      2.031997     10.80331     15.728812      0.310156   \n",
       "min        13.000000      0.000000     16.00000      1.000000      0.000000   \n",
       "25%     93526.750000      1.000000     24.00000     10.000000      0.000000   \n",
       "50%    187193.500000      3.000000     29.00000     18.000000      0.000000   \n",
       "75%    279984.250000      5.000000     37.00000     28.000000      0.000000   \n",
       "max    373662.000000      6.000000    101.00000    325.000000      1.000000   \n",
       "\n",
       "       used_premium_feature      enrolled         liked      location  \\\n",
       "count          50000.000000  50000.000000  50000.000000  50000.000000   \n",
       "mean               0.172020      0.497000      0.165000      0.517760   \n",
       "std                0.377402      0.499996      0.371184      0.499689   \n",
       "min                0.000000      0.000000      0.000000      0.000000   \n",
       "25%                0.000000      0.000000      0.000000      0.000000   \n",
       "50%                0.000000      0.000000      0.000000      1.000000   \n",
       "75%                0.000000      1.000000      0.000000      1.000000   \n",
       "max                1.000000      1.000000      1.000000      1.000000   \n",
       "\n",
       "       Institutions  ...  SecurityModal   ResendToken  TransactionList  \\\n",
       "count  50000.000000  ...   50000.000000  50000.000000     50000.000000   \n",
       "mean       0.293660  ...       0.014220      0.013340         0.013400   \n",
       "std        0.455443  ...       0.118398      0.114727         0.114981   \n",
       "min        0.000000  ...       0.000000      0.000000         0.000000   \n",
       "25%        0.000000  ...       0.000000      0.000000         0.000000   \n",
       "50%        0.000000  ...       0.000000      0.000000         0.000000   \n",
       "75%        1.000000  ...       0.000000      0.000000         0.000000   \n",
       "max        1.000000  ...       1.000000      1.000000         1.000000   \n",
       "\n",
       "       NetworkFailure    ListPicker         Other   SavingCount      CMCount  \\\n",
       "count    50000.000000  50000.000000  50000.000000  50000.000000  50000.00000   \n",
       "mean         0.008200      0.007580      6.214260      0.365020      0.92776   \n",
       "std          0.090183      0.086733      3.672561      1.405511      1.21751   \n",
       "min          0.000000      0.000000      0.000000      0.000000      0.00000   \n",
       "25%          0.000000      0.000000      3.000000      0.000000      0.00000   \n",
       "50%          0.000000      0.000000      6.000000      0.000000      0.00000   \n",
       "75%          0.000000      0.000000      8.000000      0.000000      1.00000   \n",
       "max          1.000000      1.000000     35.000000     10.000000      5.00000   \n",
       "\n",
       "            CCCount    LoansCount  \n",
       "count  50000.000000  50000.000000  \n",
       "mean       0.176860      0.788400  \n",
       "std        0.612787      0.677462  \n",
       "min        0.000000      0.000000  \n",
       "25%        0.000000      0.000000  \n",
       "50%        0.000000      1.000000  \n",
       "75%        0.000000      1.000000  \n",
       "max        3.000000      3.000000  \n",
       "\n",
       "[8 rows x 49 columns]"
      ]
     },
     "execution_count": 169,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 170,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['user', 'dayofweek', 'hour', 'age', 'numscreens', 'minigame',\n",
       "       'used_premium_feature', 'enrolled', 'liked', 'location', 'Institutions',\n",
       "       'VerifyPhone', 'BankVerification', 'VerifyDateOfBirth', 'ProfilePage',\n",
       "       'VerifyCountry', 'Cycle', 'idscreen', 'Splash', 'RewardsContainer',\n",
       "       'EditProfile', 'Finances', 'Alerts', 'Leaderboard', 'VerifyMobile',\n",
       "       'VerifyHousing', 'RewardDetail', 'VerifyHousingAmount',\n",
       "       'ProfileMaritalStatus', 'ProfileChildren ', 'ProfileEducation',\n",
       "       'ProfileEducationMajor', 'Rewards', 'AccountView', 'VerifyAnnualIncome',\n",
       "       'VerifyIncomeType', 'ProfileJobTitle', 'Login',\n",
       "       'ProfileEmploymentLength', 'WebView', 'SecurityModal', 'ResendToken',\n",
       "       'TransactionList', 'NetworkFailure', 'ListPicker', 'Other',\n",
       "       'SavingCount', 'CMCount', 'CCCount', 'LoansCount'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 170,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dataset.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 171,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset.to_csv(\"new_appdata10.csv\", index = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 172,
   "metadata": {},
   "outputs": [],
   "source": [
    "dataset = pd.read_csv(\"new_appdata10.csv\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Model Preparation"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 173,
   "metadata": {},
   "outputs": [],
   "source": [
    "y = dataset[\"enrolled\"]\n",
    "X = dataset.drop(columns = \"enrolled\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 174,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import train_test_split\n",
    "X_train, X_test, y_train, y_test = train_test_split(X, y, test_size = 0.25, random_state = 0)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 175,
   "metadata": {},
   "outputs": [],
   "source": [
    "# we do not need user ID in our model training\n",
    "# but we need to associate our model predictions to their\n",
    "# respective user IDs\n",
    "train_identifier = X_train[\"user\"]\n",
    "X_train = X_train.drop(columns = \"user\")\n",
    "test_identifier = X_test[\"user\"]\n",
    "X_test = X_test.drop(columns = \"user\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 176,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.preprocessing import StandardScaler\n",
    "sc_X = StandardScaler()\n",
    "# to preserve indexation and column names\n",
    "X_train_scaled = pd.DataFrame(sc_X.fit_transform(X_train)) \n",
    "X_test_scaled = pd.DataFrame(sc_X.transform(X_test))\n",
    "\n",
    "# recuperate column names\n",
    "X_train_scaled.columns = X_train.columns.values\n",
    "X_test_scaled.columns = X_train.columns.values\n",
    "\n",
    "# recuperate column indices\n",
    "X_train_scaled.index = X_train.index.values\n",
    "X_test_scaled.index = X_test.index.values\n",
    "\n",
    "# assign the scaled sets to our initial training and testing sets\n",
    "X_train = X_train_scaled\n",
    "X_test = X_test_scaled"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Building and training our model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 177,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
       "                   intercept_scaling=1, l1_ratio=None, max_iter=100,\n",
       "                   multi_class='auto', n_jobs=None, penalty='l2',\n",
       "                   random_state=0, solver='lbfgs', tol=0.0001, verbose=0,\n",
       "                   warm_start=False)"
      ]
     },
     "execution_count": 177,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.linear_model import LogisticRegression\n",
    "# We build an L2-regularized model\n",
    "# L2 or Ridge Regression = adds a penalty term to the loss function\n",
    "# Penalty term is proportional to squared Euclidean magnitude of weight vector\n",
    "# Ultimately prevents overfitting of the model\n",
    "classifier = LogisticRegression(random_state = 0, \n",
    "                               penalty = \"l2\")\n",
    "classifier.fit(X_train, y_train)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 178,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_pred = classifier.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 179,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.metrics import confusion_matrix, classification_report, accuracy_score\n",
    "cm = confusion_matrix(y_test, y_pred)\n",
    "cr = classification_report(y_test, y_pred)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 180,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.77      0.77      0.77      6289\n",
      "           1       0.77      0.77      0.77      6211\n",
      "\n",
      "    accuracy                           0.77     12500\n",
      "   macro avg       0.77      0.77      0.77     12500\n",
      "weighted avg       0.77      0.77      0.77     12500\n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(cr)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 181,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Test Data Accuracy: 77.00%\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.heatmap(cm, annot = True)\n",
    "print(\"Test Data Accuracy: {:.2f}%\" .format(100*accuracy_score(y_test, y_pred)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 182,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.model_selection import cross_val_score\n",
    "accuracies = cross_val_score(estimator = classifier,\n",
    "                            X = X_train,\n",
    "                            y = y_train,\n",
    "                            cv = 10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 183,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "10-fold cross-validated accuracy: 0.767 % +/- 0.012\n"
     ]
    }
   ],
   "source": [
    "print(\"10-fold cross-validated accuracy: %0.3f %% +/- %0.3f\" % (accuracies.mean(), 2*accuracies.std()))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Using our model to identify customers that would respond best to additional advertising"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "final_results = pd.concat([y_test, test_identifier], axis = 1).dropna()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "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>user</th>\n",
       "      <th>enrolled</th>\n",
       "      <th>predicted_result</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>239786</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>279644</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>98290</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>170150</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>237568</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12495</th>\n",
       "      <td>105993</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12496</th>\n",
       "      <td>288268</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12497</th>\n",
       "      <td>12780</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12498</th>\n",
       "      <td>299091</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12499</th>\n",
       "      <td>249945</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>12500 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         user  enrolled  predicted_result\n",
       "0      239786         1                 1\n",
       "1      279644         1                 1\n",
       "2       98290         0                 0\n",
       "3      170150         1                 1\n",
       "4      237568         1                 1\n",
       "...       ...       ...               ...\n",
       "12495  105993         0                 1\n",
       "12496  288268         1                 1\n",
       "12497   12780         1                 1\n",
       "12498  299091         1                 0\n",
       "12499  249945         0                 0\n",
       "\n",
       "[12500 rows x 3 columns]"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "final_results[\"predicted_result\"] = y_pred\n",
    "final_results[[\"user\", \"enrolled\", \"predicted_result\"]].reset_index(drop = True)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.6"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}