Customer_Churn_Analysis.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<br>\n",
    "\n",
    "<span style=\"font-size:2em;font-weight:bold\">Telco Customer Churn Analysis</span>\n",
    "\n",
    "In this exercise, we'll be creating a simple [Flask application](http://dachurn.herokuapp.com/) using Telco Customer Churn dataset retrieved from [IBM Cognos Analytic](https://community.ibm.com/community/user/businessanalytics/blogs/steven-macko/2019/07/11/telco-customer-churn-1113). Full data glossary can be accessed [here](https://github.com/ttnsy/dac-telco#dataset).\n",
    "\n",
    "This notebook contains guidances & tasks on the data processing for the application."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<span style=\"font-size:1.5em;font-weight:bold\">Libraries & Setup</span>:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import matplotlib.ticker as mtick"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Load & Pre-process data\n",
    "\n",
    "Let's begin by reading our dataset in. The data will be using is `telcochurn.csv` which located under `data/` directory:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "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>customer_id</th>\n",
       "      <th>city</th>\n",
       "      <th>gender</th>\n",
       "      <th>senior_citizen</th>\n",
       "      <th>dependents</th>\n",
       "      <th>tenure_months</th>\n",
       "      <th>phone_service</th>\n",
       "      <th>internet_service</th>\n",
       "      <th>contract</th>\n",
       "      <th>monthly_charges</th>\n",
       "      <th>total_charges</th>\n",
       "      <th>churn_label</th>\n",
       "      <th>cltv</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3668-QPYBK</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>2</td>\n",
       "      <td>Single Line</td>\n",
       "      <td>DSL</td>\n",
       "      <td>Month-to-month</td>\n",
       "      <td>53.85</td>\n",
       "      <td>108.15</td>\n",
       "      <td>Yes</td>\n",
       "      <td>3239</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9237-HQITU</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>2</td>\n",
       "      <td>Single Line</td>\n",
       "      <td>Fiber optic</td>\n",
       "      <td>Month-to-month</td>\n",
       "      <td>70.70</td>\n",
       "      <td>151.65</td>\n",
       "      <td>Yes</td>\n",
       "      <td>2701</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9305-CDSKC</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>8</td>\n",
       "      <td>Multiple Lines</td>\n",
       "      <td>Fiber optic</td>\n",
       "      <td>Month-to-month</td>\n",
       "      <td>99.65</td>\n",
       "      <td>820.50</td>\n",
       "      <td>Yes</td>\n",
       "      <td>5372</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7892-POOKP</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>28</td>\n",
       "      <td>Multiple Lines</td>\n",
       "      <td>Fiber optic</td>\n",
       "      <td>Month-to-month</td>\n",
       "      <td>104.80</td>\n",
       "      <td>3046.05</td>\n",
       "      <td>Yes</td>\n",
       "      <td>5003</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0280-XJGEX</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>49</td>\n",
       "      <td>Multiple Lines</td>\n",
       "      <td>Fiber optic</td>\n",
       "      <td>Month-to-month</td>\n",
       "      <td>103.70</td>\n",
       "      <td>5036.30</td>\n",
       "      <td>Yes</td>\n",
       "      <td>5340</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  customer_id         city  gender senior_citizen dependents  tenure_months  \\\n",
       "0  3668-QPYBK  Los Angeles    Male             No         No              2   \n",
       "1  9237-HQITU  Los Angeles  Female             No        Yes              2   \n",
       "2  9305-CDSKC  Los Angeles  Female             No        Yes              8   \n",
       "3  7892-POOKP  Los Angeles  Female             No        Yes             28   \n",
       "4  0280-XJGEX  Los Angeles    Male             No        Yes             49   \n",
       "\n",
       "    phone_service internet_service        contract  monthly_charges  \\\n",
       "0     Single Line              DSL  Month-to-month            53.85   \n",
       "1     Single Line      Fiber optic  Month-to-month            70.70   \n",
       "2  Multiple Lines      Fiber optic  Month-to-month            99.65   \n",
       "3  Multiple Lines      Fiber optic  Month-to-month           104.80   \n",
       "4  Multiple Lines      Fiber optic  Month-to-month           103.70   \n",
       "\n",
       "   total_charges churn_label  cltv  \n",
       "0         108.15         Yes  3239  \n",
       "1         151.65         Yes  2701  \n",
       "2         820.50         Yes  5372  \n",
       "3        3046.05         Yes  5003  \n",
       "4        5036.30         Yes  5340  "
      ]
     },
     "execution_count": 75,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "telco = pd.read_csv('data/telcochurn.csv')\n",
    "telco.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A very common step on your exploratory process is to make sure that all your columns have stored in the right data types. In the following cell, use `dtypes` to acquire information of data type of each column:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Your code here\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Notice that all categorical informations from the data were stored as `object` rather than `category`.\n",
    "\n",
    "## Adjust Data Types\n",
    "\n",
    "**Task 1**: Fill in the blank (`___`) part of the code below to cast all `object` to `category`:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "catcol = telco.select_dtypes('object').columns\n",
    "telco[catcol] = telco[catcol].apply(lambda x: x.astype('_______'))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Categorize Tenure\n",
    "\n",
    "The `tenure_months` column indicates the total amount of months that the customer has been with the company by the end of the quarter. Your next task is to create a new column named `tenure_group` which categorize the tenure months by the following conditions:\n",
    "- **< 1 Year**: 12 tenure months and below\n",
    "- **1-2 Year**: More than 12 tenure months up until 24\n",
    "- **2-4 Year**: More than 24 tenure months up until 48\n",
    "- **4-5 Year**: More than 48 tenure months up until 60\n",
    "- **\\> 5 Year**: More than 60 tenure months\n",
    "\n",
    "**Task 2**: Fill in the blank (`___`) part to create the `tenure_group` column!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Tenure Months to grouping categories\n",
    "def grouping_tenure(telco) :\n",
    "    if telco[\"_______\"] <= 12 :\n",
    "        return \"< 1 Year\"\n",
    "    elif (telco[\"_______\"] > 12) & (telco[\"_______\"] __ _______ ):\n",
    "        return \"1-2 Year\"\n",
    "    elif (telco[\"_______\"] __ _______) & (telco[\"_______\"] <= 48) :\n",
    "        return \"2-4 Year\"\n",
    "    elif (telco[\"_______\"] __ _______) & (telco[\"_______\"] __ _______) :\n",
    "        return \"4-5 Year\"\n",
    "    else:\n",
    "        return \"> 5 Year\"\n",
    "\n",
    "telco[\"tenure_group\"] = telco.apply(lambda telco: _______(telco), axis = 1) \n",
    "\n",
    "# Adjust category order\n",
    "tenure_group = [\"< 1 Year\", \"1-2 Year\", \"2-4 Year\", \"4-5 Year\", \"> 5 Year\"]\n",
    "telco[\"tenure_group\"] = pd.Categorical(telco[\"tenure_group\"], categories = tenure_group, ordered=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Wrap as Function\n",
    "\n",
    "Lastly, let's wrap the whole process above as a function named `load_telco()`. This process is essential because in your Flask application later, you will need to define all your data processing as specific functions. \n",
    "\n",
    "**Task 3**: Fill in the blank (`___`) part with the same process you have done previously to complete the function."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "metadata": {},
   "outputs": [],
   "source": [
    "def load_telco():\n",
    "    # Read data\n",
    "    telco = pd.read_csv('data/telcochurn.csv')\n",
    "    \n",
    "    # Adjust dtypes\n",
    "    catcol = telco.select_dtypes('object').columns\n",
    "    telco[catcol] = telco[catcol].apply(lambda x: x.astype('______'))\n",
    "    \n",
    "    # Tenure Months to grouping categories\n",
    "    def grouping_tenure(telco) :\n",
    "        if telco[\"_______\"] <= 12 :\n",
    "            return \"< 1 Year\"\n",
    "        elif (telco[\"_______\"] > 12) & (telco[\"_______\"] __ ____ ):\n",
    "            return \"1-2 Year\"\n",
    "        elif (telco[\"_______\"] __ ____) & (telco[\"_______\"] <= 48) :\n",
    "            return \"2-4 Year\"\n",
    "        elif (telco[\"_______\"] __ ____) & (telco[\"_______\"] __ ____) :\n",
    "            return \"4-5 Year\"\n",
    "        else:\n",
    "            return \"> 5 Year\"\n",
    "        \n",
    "    telco[\"tenure_group\"] = telco.apply(lambda telco: _______(telco), axis = 1)\n",
    "    \n",
    "    # Adjust category order\n",
    "    tenure_group = [\"< 1 Year\", \"1-2 Year\", \"2-4 Year\", \"4-5 Year\", \"> 5 Year\"]\n",
    "    telco[\"tenure_group\"] = pd.Categorical(telco[\"tenure_group\"], categories = tenure_group, ordered=True)\n",
    "    \n",
    "    return(telco)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Congratulations, you have done the very first challenge of this capstone project!\n",
    "\n",
    "If you take a look at the project directory, you will find two separate scripts; `app.py` and `helper.py`. `app.py` handles the routing and within `helper.py` you would define function to be called in `app`.\n",
    "\n",
    "If you open up the `helper.py`, you will find all the functions needed to create the dashboard contents such as plots and data values. You may now copy the `load_telco()` function definition above and overwrite the existing function on the script and save it.\n",
    "\n",
    "In `app.py`, notice that there's a line of `data = load_telco()`. that you've save your function in separate python file/script, you can import the function to another script as the following:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "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>customer_id</th>\n",
       "      <th>city</th>\n",
       "      <th>gender</th>\n",
       "      <th>senior_citizen</th>\n",
       "      <th>dependents</th>\n",
       "      <th>tenure_months</th>\n",
       "      <th>phone_service</th>\n",
       "      <th>internet_service</th>\n",
       "      <th>contract</th>\n",
       "      <th>monthly_charges</th>\n",
       "      <th>total_charges</th>\n",
       "      <th>churn_label</th>\n",
       "      <th>cltv</th>\n",
       "      <th>tenure_group</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3668-QPYBK</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>No</td>\n",
       "      <td>2</td>\n",
       "      <td>Single Line</td>\n",
       "      <td>DSL</td>\n",
       "      <td>Month-to-month</td>\n",
       "      <td>53.85</td>\n",
       "      <td>108.15</td>\n",
       "      <td>Yes</td>\n",
       "      <td>3239</td>\n",
       "      <td>&lt; 1 Year</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>9237-HQITU</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>2</td>\n",
       "      <td>Single Line</td>\n",
       "      <td>Fiber optic</td>\n",
       "      <td>Month-to-month</td>\n",
       "      <td>70.70</td>\n",
       "      <td>151.65</td>\n",
       "      <td>Yes</td>\n",
       "      <td>2701</td>\n",
       "      <td>&lt; 1 Year</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>9305-CDSKC</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>8</td>\n",
       "      <td>Multiple Lines</td>\n",
       "      <td>Fiber optic</td>\n",
       "      <td>Month-to-month</td>\n",
       "      <td>99.65</td>\n",
       "      <td>820.50</td>\n",
       "      <td>Yes</td>\n",
       "      <td>5372</td>\n",
       "      <td>&lt; 1 Year</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>7892-POOKP</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>Female</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>28</td>\n",
       "      <td>Multiple Lines</td>\n",
       "      <td>Fiber optic</td>\n",
       "      <td>Month-to-month</td>\n",
       "      <td>104.80</td>\n",
       "      <td>3046.05</td>\n",
       "      <td>Yes</td>\n",
       "      <td>5003</td>\n",
       "      <td>2-4 Year</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0280-XJGEX</td>\n",
       "      <td>Los Angeles</td>\n",
       "      <td>Male</td>\n",
       "      <td>No</td>\n",
       "      <td>Yes</td>\n",
       "      <td>49</td>\n",
       "      <td>Multiple Lines</td>\n",
       "      <td>Fiber optic</td>\n",
       "      <td>Month-to-month</td>\n",
       "      <td>103.70</td>\n",
       "      <td>5036.30</td>\n",
       "      <td>Yes</td>\n",
       "      <td>5340</td>\n",
       "      <td>4-5 Year</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  customer_id         city  gender senior_citizen dependents  tenure_months  \\\n",
       "0  3668-QPYBK  Los Angeles    Male             No         No              2   \n",
       "1  9237-HQITU  Los Angeles  Female             No        Yes              2   \n",
       "2  9305-CDSKC  Los Angeles  Female             No        Yes              8   \n",
       "3  7892-POOKP  Los Angeles  Female             No        Yes             28   \n",
       "4  0280-XJGEX  Los Angeles    Male             No        Yes             49   \n",
       "\n",
       "    phone_service internet_service        contract  monthly_charges  \\\n",
       "0     Single Line              DSL  Month-to-month            53.85   \n",
       "1     Single Line      Fiber optic  Month-to-month            70.70   \n",
       "2  Multiple Lines      Fiber optic  Month-to-month            99.65   \n",
       "3  Multiple Lines      Fiber optic  Month-to-month           104.80   \n",
       "4  Multiple Lines      Fiber optic  Month-to-month           103.70   \n",
       "\n",
       "   total_charges churn_label  cltv tenure_group  \n",
       "0         108.15         Yes  3239     < 1 Year  \n",
       "1         151.65         Yes  2701     < 1 Year  \n",
       "2         820.50         Yes  5372     < 1 Year  \n",
       "3        3046.05         Yes  5003     2-4 Year  \n",
       "4        5036.30         Yes  5340     4-5 Year  "
      ]
     },
     "execution_count": 72,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from helper import load_telco\n",
    "\n",
    "data = load_telco()\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<span style=\"font-size:2em;font-weight:bold\">Data Wrangling & Visualization</span>\n",
    "\n",
    "Now that you're data is ready, we're now ready to prepare the explanatory data in our dashboard!\n",
    "\n",
    "Within `app.py` you will see the `index()` function will return a rendered templates along with some variables:\n",
    "- `card_data`: quick summary including **Customer Churn Rate**, **Retaining Customer** & **Average Lifetime Value**\n",
    "- `plot_tenure_churn`: **Churn Rate by Tenure Group** plot in **Customer Tenure** tab\n",
    "- `plot_tenure_cltv`: **Average Lifetime Value by Tenure** plot in **Customer Tenure** tab\n",
    "- `plot_phone`: **Phone Service Customer** plot in **Subscription Service** tab\n",
    "- `plot_internet`: **Internet Service Customer** plot in **Subscription Service** tab\n",
    "\n",
    "The values and plots above were resulted from all EDA techniques you have learned in the pervious workshops. Your next challenge is to recall how to use pandas subsetting method, aggregation tables and simple visualization!\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Extract Quick Summary\n",
    "\n",
    "Let's acquire the needed information for the quick summary in `card_data`. There are 3 values you need to extract:\n",
    "- **Customer Churn Rate**: The percentage of churn customers (`churn_label` = Yes in `churn_label`)  \n",
    "- **Retaining Customer**: The percentage of retaining customers (`churn_label` = No in `churn_label`)  \n",
    "- **Average Lifetime Value**: Average customer lifetime value (`cltv`)\n",
    "\n",
    "## Churn & Retaining Percentage\n",
    "**Task 4**: Fill in the blank (`___`) part of the code below to create a proportion table of Churn & Retaining customer!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>col_0</th>\n",
       "      <th>percent</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>churn_label</th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>No</th>\n",
       "      <td>73.463013</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Yes</th>\n",
       "      <td>26.536987</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "col_0          percent\n",
       "churn_label           \n",
       "No           73.463013\n",
       "Yes          26.536987"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "table = pd.crosstab(\n",
    "    data['______'],\n",
    "    columns = 'percent',\n",
    "    normalize = ______\n",
    ")*100\n",
    "\n",
    "table"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We can extract the value from our table by using `.loc[]` subsetting method. \n",
    "\n",
    "**Task 5**: Fill in the blank (`___`) part to extract the information of churned customer!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'26.54%'"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "percent_churn = ______.loc[______, ______].round(2)\n",
    "\n",
    "f'{percent_churn}%'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Task 6**: Use the same method as above to also extract the information of retaining customer!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'73.46%'"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "percent_retain = _______\n",
    "\n",
    "f'{percent_retain}%'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Average Lifetime Value\n",
    "\n",
    "**Task 7**: The average lifetime value stored the average (`mean`) value of `cltv` variable. Extract the value on the cell code below!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Your code here\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<span style=\"font-size:1.5em;font-weight:bold\">Update the application scripts!</span>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Lastly, follow these instructions to complete the next part of your application scripts:\n",
    "- **On helper.py:**\n",
    "    - Use the code in **Task 4** to complete `table_churn()` function in helper.\n",
    "\n",
    "\n",
    "- **On app.py:**\n",
    "    - Add table_churn function when importing from helper.<br> (To import multiple functions, you can simply use comma such as: `from helper import load_telco, table_churn`)\n",
    "    - \"Uncomment\" the codes under `# generate value for cards` section then use the code in **Task 5** & **Task 6** to complete the blank (`___`) part of the line\n",
    "    - Lastly, in `render_template()` at the end of `index()` in `app.py`, change `card_data=None` to `card_data=card_data`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Generate Plot\n",
    "\n",
    "## Subscription Service\n",
    "\n",
    "The \"Subscription Service\" tab contains explanatory analysis on the proportion of churned-retained customers in each phone and internet service. There are two plots you need to recreate for this tab; **Phone Service Customer** and **Internet Service Customer**"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Phone Service Customer\n",
    "\n",
    "The Phone Service Customer plot compares the frequency proportion of churned and retained customers in each service of the phone subscription.\n",
    "\n",
    "If you use `data['phone_service'].unique()`, you'll find that apparently there are two phone services offered by the company; `Single Line` and `Multiple Lines`while `No` means the particular customer is not subscribed to the company's phone service."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Your code here\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Task 8**:  \n",
    "Use `pd.crosstab()` to get the frequency proportion, then recall how to use `.plot(kind = 'barh')` to create a horizontal bar plot. You can also enhance the visualization such as adjusting the color by add the color parameter to your plot function call `.plot(kind = 'bar',color=['#53a4b1','#c34454'])`.\n",
    "\n",
    "For now, **do not mind the plot's customization** as you can copy-and-paste the codes defined under the `plot_phone()` function in `helper.py`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>churn_label</th>\n",
       "      <th>No</th>\n",
       "      <th>Yes</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>phone_service</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Multiple Lines</th>\n",
       "      <td>2121</td>\n",
       "      <td>850</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>No</th>\n",
       "      <td>512</td>\n",
       "      <td>170</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Single Line</th>\n",
       "      <td>2541</td>\n",
       "      <td>849</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "churn_label       No  Yes\n",
       "phone_service            \n",
       "Multiple Lines  2121  850\n",
       "No               512  170\n",
       "Single Line     2541  849"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Your code here\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Phone Service Customer')"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhgAAAF1CAYAAACqBpMJAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+j8jraAAAfZ0lEQVR4nO3debhddX3v8fcnAyQBkSEIOBGIyuTDYCKIQM3VOhQhIGKLchGqFdGrV9ti1eItcbxawCLKvYrgdUJQQBFbK6DFlqEqiSaMyqBQCIKiRZOKmJDv/WOvg5vjOck54ZfsDO/X8+znrL2G3/qu39nZ65PfWvvsVBWSJEktTRh0AZIkacNjwJAkSc0ZMCRJUnMGDEmS1JwBQ5IkNWfAkCRJzRkwpPVckm8n+YtB17E2JVmaZOdB1yFpdAYMaT2Q5I4kD3Yn1vuSfDrJ5oOuq1+Sw5IsTPLrJPcn+ZckO62JfVXV5lX149btJnlVkvldP/80yT8nOfAxtjkvyedb1SitLwwY0vrj0KraHHgWMBt414DreUSSpwGfBf4aeDywE3Am8PBqtDWpbXVj3u9fAacDHwC2A54K/B/gsEHUsyYMqm+1cTJgSOuZqloM/DPwzL7ZOya5OsmSJJclmT60IMncJDcmeaC7nLJb37I7kpyY5Lokv0ryxSRT+pYf0o1KPJDkmiR7jlLW3sBPqupb1bOkqi6qqv/o2pmQ5B1Jbk/yiyRfSrJ1t2xGkkry2iT/AfxLN3Lwpv4dJFmU5IhuurpQQ5KpSU5Lcmd3DFclmdote05X9wPd9nNGKj7J44H3AP+jqr5cVf9VVcuq6mtV9bZunU8neV/fNnOS3N33/O1JFne/gx8leUGSlwB/C/xZNyqyqFv3iUkuSfLLJLcleV1fO/OSXJDk811b1yd5RpJ3JvlZkruSvKi/9iTndCMui5O8L8nEbtlx3eviH5L8Apg3yu9Pas6AIa1nkjwFOBj4Qd/sVwF/DjwB2AQ4sVv3GcB5wFuBbYGvA19Lsknftn8KvITeqMOewHHdtvsAnwJeD2wDfAK4JMmmI5T1fWDX7kT230a4fPNm4HDgecATgf+kN8LR73nAbsCLu5pf2XfMuwM7Av80wr5PBWYBzwW2Bv4GWJHkSd367+vmnwhclGTbEdrYH5gCfGWEZauUZBfgTcCzq+px3THcUVXfoDci8sXuss5e3SbnA3fT64sjgQ8keX5fk4cCnwO2ovd7vpTe+/WT6AWhT/St+2lgOfA0YB/gRUD/PTn7AT+mNyrz/tU5Pml1GDCk9cfFSR4ArgL+ld6Ja8j/q6pbqupB4Ev0RhQA/gz4p6q6vKqW0TsZT6V3Mh5yRlXdU1W/BL7Wt+3xwCeq6rtV9XBVfQZ4CHjO8MK6+yHm0DsBfgm4f9h9IicAJ1XV3VX1EL3/SR85bMh+Xjdy8CC9E/3eSXbslh0NfLnb9hFJJgCvAd5SVYu7Oq/p1vvvwNer6utVtaKqLgfm0wtnw20D3F9Vy0dYNhYPA5sCuyeZXFV3VNXtI63YBcQDgLdX1W+raiFwNvDqvtWurKpLu3ouoBcOP9j9Ds8HZiTZMsl23fG8teu7nwH/ABzV19Y9VfXRqlre9a20VhgwpPXH4VW1ZVXtWFVvHHayuLdv+jfA0In9icCdQwuqagVwF70gsKptdwT+uru88EAXbp7StfkHquo7VfWnVbUtcBDwR8BJfW19pa+dm+mdlLfra+KuvraW0Bt9GDpRvhI4d4TdTqc38jDSyXxH4BXD6j8Q2GGEdX8BTF/dexSq6jZ6o0TzgJ8lOT/JiP1Er/9+2R3jkDt59O/kvr7pB+mFn4f7nkPv97QjMBn4ad8xfoLeSNaQu5AGwIAhbdjuoXcSAiBJ6IWExWPY9i7g/V2oGXpMq6rzVrVhVV0LfJnf3ydyF/Anw9qa0t1P8shmw5o5D3hlkqHLF1eMsKv7gd8CM0ep/3PD9rlZVX1whHX/nd7ozOErOaz/Aqb1Pd++f2FVfaGqDqTX3wV8aJTjugfYOsnj+uY9lbH9Toa7q6t7et8xblFVe/SXthrtSo+ZAUPasH0JeGl3w+Fkep/yeAi4ZgzbfhI4Icl+6dksyUuHnRgBSHJgktcleUL3fFdgLvCdbpWPA+8fuuSRZNskq/p0xtfpnazfQ+8ehhXDV+jmfQr4cHfj5MQk+3f3iXweODTJi7v5U7obM588Qju/Av4OODPJ4UmmJZmc5E+S/H232kLg4CRbJ9me3ojF0PHvkuT53X5/S2+UYaje++hd0pjQ7esuev3/v7ua9gRe29U7LlX1U+Ay4LQkW6R3M+3MJM8bb1tSawYMaQNWVT+idy/CR+n9b/9Qeh93/d0Ytp0PvA74GL2bMm+juwF0BA/QCxTXJ1kKfIPefRRDJ+ePAJcAlyVZQi947LeK/T9EbxTkj4EvrGTVE4HrgWuBX9IbOZjQncgPo/cpjp/T+9/+2xjlfa+qTgP+it7Hf4fWfxNwcbfK54BFwB30Tupf7Nt8U+CD9Pr4XnqXKN7ZLbug+/mLJN/vpl8JzKA3mvEV4OSq+uZKjnFlXk3vxt6b6P2eLmTky0DSWpUqR88kSVJbjmBIkqTmDBiSJKk5A4YkSWrOgCFJkpozYEiSpOb8Zr2Gpk+fXjNmzBh0GZIkrTULFiy4v/sLvo9iwGhoxowZzJ8/f9BlSJK01iS5c6T5XiKRJEnNGTAkSVJzBgxJktSc92BIkjSK3/3ud9x+++385je/GXQpAzdt2jRmzpzJJptsMqb1DRiSJI3i9ttvZ8stt2SXXXZhwoSNd9B/xYoV3HvvvSxcuJBdd92VLbbYYpXbbLy9JUnSKvzmN79hu+2226jDBcCECRPYfvvtmTBhAhdddBHLli1b9TZroS5JktZbG3u4GDJhwgSSsGTJEn7961+vev21UJMkSVpNEydOZO+99+aZz3wmhx56KA888MBK17/44ou56aabVtnuxz/+cT772c+Ou54krFixYpXreQ+GJEljdOS5lzRt78Kj565ynalTp7Jw4UIAjj32WM4880xOOumkUde/+OKLOeSQQ9h9991X2u4JJ5wwvmLHyREMSZLWE/vvvz+LFy8GejegvuQlL2HWrFkcdNBB/PCHP+Saa67hkksu4W1vext77703t99+O5/85Cd59rOfzV577cXLX/7yRz4RM2/ePE499VQA5syZw9vf/nb23XdfnvGMZ3DllVc+5loNGJIkrQcefvhhvvWtbzF3bm/U4/jjj+ejH/0oCxYs4NRTT+WNb3wjz33uc5k7dy6nnHIKCxcuZObMmRxxxBFce+21LFq0iN12241zzjlnxPaXL1/O9773PU4//XTe/e53P+Z6vUQiSdI67MEHH2Tvvfdm8eLF7LbbbrzwhS9k6dKlXHPNNbziFa94ZL2HHnpoxO1vuOEG3vWud/HAAw+wdOlSXvziF4+43hFHHAHArFmzuOOOOx5z3Y5gSJK0Dhu6B+POO++kqjjzzDNZsWIFW265JQsXLnzkcfPNN4+4/XHHHcfHPvYxrr/+ek4++WR++9vfjrjepptuCvRuKl2+fPljrtuAIUnSemDatGmcccYZnHbaaUybNo2ddtqJCy64AICqYtGiRQA87nGPY8mSJY9st2TJEnbYYQeWLVvGueeeu9bqNWBIkrSe2Geffdhzzz0577zzOPfccznnnHPYa6+92GOPPfjqV78KwFFHHcUpp5zCPvvsw+2338573/te9ttvPw444AB23XXXtVZrqmqt7WxDN3v27Jo/f/6gy5AkNbJgwQJmzZo16DLWGQsWLOCaa67hqKOOYttttwUgyYKqmj18XUcwJElScwYMSZLUnAFDkiQ1Z8CQJEnNGTAkSVJzBgxJktScAUOSpHXYvffey1FHHcXMmTOZNWsWBx98MGeddRaHHHLIoEtbKb+LRJKkMbrqha9Y9UrjcODlF6x0eVXxspe9jGOPPZbzzz8fgEWLFnHJJY/ta+OXL1/OpElrNgI4giFJ0jrqiiuuYPLkyZxwwgmPzNtrr7046KCDWLp0KUceeSS77rorRx99NEN/OHPGjBncf//9AMyfP585c+YAva9nP+aYYzjggAM45phjmDdvHq95zWuYM2cOO++8M2eccUbT2h3BaGjpLT9unm614VvV/2AkbbxuuOGGUf+S6A9+8ANuvPFGnvjEJ3LAAQdw9dVXc+CBB660vZtuuomrrrqKqVOnMm/ePH74wx9yxRVXsGTJEnbZZRfe8IY3MHny5Ca1O4IhSdJ6aN999+XJT34yEyZMYO+99x7TV6zPnTuXqVOnPvL8pS99KZtuuinTp0/nCU94Avfdd1+z+gwYkiSto/bYYw8WLFgw4rKhr1eHR3/F+qRJk1ixYgXAH3w1+2abbTamNlowYEiStI56/vOfz0MPPcRZZ531yLzrrruOK6+8ctRtZsyY8Ugoueiii9Z4jaMxYEiStI5Kwle+8hW++c1vMnPmTPbYYw/e+c53sv3224+6zcknn8xb3vIWZs+ezcSJE9ditY/m17U3tOsWW9fZ+71g0GVoPeNNntK6y69rfzS/rl2SJA2UAUOSJDVnwJAkSc0ZMCRJWomhj3xu7MbbDwYMSZJGMW3aNO69996NPmSsWLGCe++9l2XLllFVJFnlNv6pcEmSRjFz5kxuvvlm7rnnnjGdVDdky5Yt49Zbb2XChAl/8Ae7RmLAkCRpFJtssgl77rknV199NfPnzyfJRh00Jk6cyCGHHPKoPzc+GgOGJEkrkYQDDzyQ2bNns2zZskGXM1BTpkwZ85ehGTAkSRqDKVOmMGXKlEGXsd7wJk9JktScAUOSJDVnwJAkSc0ZMCRJUnMGDEmS1JwBQ5IkNWfAkCRJzRkwJElScwYMSZLU3BoLGElOSnJjkuuSLEyyXzf/7CS7r2abM5LcMM5tlo4w74Qkr16dGiRJ0qqtkT8VnmR/4BDgWVX1UJLpwCYAVfUXa2Kf41FVHx90DZIkbcjW1AjGDsD9VfUQQFXdX1X3ACT5dpLZ3fTSJO9PsijJd5Js182f2T2/Psn7RhmFmJjklCTXdqMkrx9rcUnmJTmxr54PJflekluSHPRY25ckaWO3pr7s7DLg75LcAnwT+GJV/esI620GfKeqTkry98DrgPcBHwE+UlXnJTlhlH28FvhVVT07yabA1Ukuq6qfrEa9k6pq3yQHAycDfzzW9pMcDxwPMHWbbTn9uGNWY/famJ1+7iWDLkHSRuDCo+eu1f2tkRGMqloKzKJ34v058MUkx42w6u+Af+ymFwAzuun9gQu66S+MspsXAa9OshD4LrAN8PTVLPnLI9Qwpvar6qyqml1VszfdYovV3L0kSRuWNfZ17VX1MPBt4NtJrgeOBT49bLVlVVXd9MPjrCfAm6vq0sdYKsBDI9TQsn1JkjYqa2QEI8kuSfr/t783cOc4mvgO8PJu+qhR1rkUeEOSyd0+n5Fks3EXO7o13b4kSRusNTWCsTnw0SRbAsuB2+juUxijtwKfT3IS8A3gVyOscza9yxnfTxJ6l2IOH2G9aUnu7nv+4THWMNb2JUnSMPn9FYp1R5JpwINVVUmOAl5ZVYcNuq5V2Wrnp9UL3jvW/CJJ0tqzpm7yTLKgqmYPn7/G7sF4jGYBH+tGDh4AXjPgeiRJ0jiskwGjqq4E9hp0HZIkafX4XSSSJKk5A4YkSWrOgCFJkpozYEiSpOYMGJIkqTkDhiRJas6AIUmSmjNgSJKk5gwYkiSpOQOGJElqzoAhSZKaM2BIkqTmDBiSJKk5A4YkSWrOgCFJkpozYEiSpOYMGJIkqTkDhiRJas6AIUmSmps06AI2JDO33pILj5476DIkSRo4RzAkSVJzBgxJktScAUOSJDVnwJAkSc0ZMCRJUnMGDEmS1JwBQ5IkNWfAkCRJzRkwJElScwYMSZLUnAFDkiQ1Z8CQJEnNGTAkSVJzBgxJktScAUOSJDVnwJAkSc0ZMCRJUnMGDEmS1JwBQ5IkNWfAkCRJzRkwJElScwYMSZLUnAFDkiQ1Z8CQJEnNGTAkSVJzBgxJktScAUOSJDVnwJAkSc0ZMCRJUnMGDEmS1JwBQ5IkNWfAkCRJzRkwJElScwYMSZLUnAFDkiQ1Z8CQJEnNGTAkSVJzBgxJktScAUOSJDVnwJAkSc0ZMCRJUnMGDEmS1JwBQ5IkNWfAkCRJzRkwJElScwYMSZLUnAFDkiQ1Z8CQJEnNGTAkSVJzBgxJktScAUOSJDVnwJAkSc0ZMCRJUnMGDEmS1NykQRewIVl6y4+56oWvWKv7PPDyC9bq/iRJGgtHMCRJUnMGDEmS1JwBQ5IkNWfAkCRJzRkwJElScwYMSZLUnAFDkiQ1Z8CQJEnNGTAkSVJzBgxJktScAUOSJDVnwJAkSc0ZMCRJUnMGDEmS1JwBQ5IkNWfAkCRJzRkwJElScwYMSZLUnAFDkiQ1Z8CQJEnNGTAkSVJzBgxJktTcRh8wklSS0/qen5hk3gBLkiRpvbfRBwzgIeCIJNMHXYgkSRsKAwYsB84C/nL4giQzkvxLkuuSfCvJU9d+eZIkrX8MGD1nAkcnefyw+R8FPlNVewLnAmes9cokSVoPpaoGXcNAJVlaVZsneQ+wDHgQ2Lyq5iW5H9ihqpYlmQz8tKqmD9v+eOB4gKnbbDvr4I+cvbYPYYN34dFzB12CJGkUSRZU1ezh8x3B+L3TgdcCm41no6o6q6pmV9XsTbfYYs1UJknSesaA0amqXwJfohcyhlwDHNVNHw1cubbrkiRpfWTAeLTTgP5LIG8G/jzJdcAxwFsGUpUkSeuZSYMuYNCqavO+6fuAaX3P7wSeP4i6JElanzmCIUmSmjNgSJKk5gwYkiSpOQOGJElqzoAhSZKaM2BIkqTmDBiSJKk5A4YkSWrOgCFJkpozYEiSpOYMGJIkqTkDhiRJas6AIUmSmjNgSJKk5gwYkiSpOQOGJElqzoAhSZKaM2BIkqTmDBiSJKk5A4YkSWrOgCFJkpozYEiSpOYmDbqADcnMrbfkwqPnDroMSZIGzhEMSZLUnAFDkiQ1Z8CQJEnNGTAkSVJzBgxJktScAUOSJDVnwJAkSc0ZMCRJUnMGDEmS1JwBQ5IkNWfAkCRJzRkwJElScwYMSZLUnAFDkiQ1Z8CQJEnNGTAkSVJzBgxJktScAUOSJDVnwJAkSc0ZMCRJUnMGDEmS1JwBQ5IkNWfAkCRJzRkwJElScwYMSZLUnAFDkiQ1Z8CQJEnNGTAkSVJzBgxJktScAUOSJDVnwJAkSc0ZMCRJUnMGDEmS1JwBQ5IkNWfAkCRJzRkwJElScwYMSZLUnAFDkiQ1Z8CQJEnNGTAkSVJzBgxJktScAUOSJDVnwJAkSc0ZMCRJUnMGDEmS1JwBQ5IkNWfAkCRJzRkwJElScwYMSZLUnAFDkiQ1Z8CQJEnNGTAkSVJzBgxJktTcpEEXsCFZesuPueqFrxh0GVoPHXj5BYMuQZKacgRDkiQ1Z8CQJEnNGTAkSVJzBgxJktScAUOSJDVnwJAkSc0ZMCRJUnMGDEmS1JwBQ5IkNWfAkCRJzRkwJElScwYMSZLUnAFDkiQ1Z8CQJEnNGTAkSVJzBgxJktScAUOSJDVnwJAkSc0ZMCRJUnMGDEmS1JwBQ5IkNWfAkCRJzRkwJElSc6sMGEkqyef7nk9K8vMk/ziGbZd2P2ckeVXf/NlJzljFtjOS3LCqfYy0v2HzTkjy6vG0I0mSHptJY1jnv4BnJplaVQ8CLwQWj3M/M4BXAV8AqKr5wPxxtrFaqurja2M/kiTp98Z6ieTrwEu76VcC5w0tSDIvyYl9z29IMmPY9h8EDkqyMMlfJpkzNALSbf+5JP+e5NYkrxu+8yQTk5yS5Nok1yV5/VgPsL++JN9O8qEk30tyS5KDVtZ+kh2S/FtX9w1D60uSpJUbywgGwPnA33WhYE/gU8B4TrbvAE6sqkMAkswZtnxP4DnAZsAPkvzTsOWvBX5VVc9OsilwdZLLquon46hhyKSq2jfJwcDJwB+P1j5wBHBpVb0/yURg2vDGkhwPHA8wdZttOf24Y1ajJG3sTj/3kkGXIG0wLjx67qBLEGMMGFV1XTcq8Up6oxmtfbW7/PJgkiuAfYGFfctfBOyZ5Mju+eOBpwOrEzC+3P1cQO/Szcravxb4VJLJwMVV1V8TAFV1FnAWwFY7P61Wox5JkjY4Yx3BALgEOBWYA2zTN385j77UMmU16hh+Yh7+PMCbq+rS1Wh7uIe6nw/z++Mftf0kf0Tv8tCnk3y4qj7boAZJkjZo4/mY6qeAd1fV9cPm3wE8CyDJs4CdRth2CfC4lbR9WJIpSbahF2CuHbb8UuAN3UgCSZ6RZLNx1L4qI7afZEfgvqr6JHA23XFKkqSVG/MIRlXdDYz00dKLgFcnuRH4LnDLCOtcBzycZBHwaeAHIyy/ApgOvLeq7hl2o+jZ9C5nfD9JgJ8Dh4+wn2lJ7u57/uGVH9Uq258DvC3JMmAp4MddJUkag1QN9raBJPOApVV16kALaWCrnZ9WL3jvWDONJGlN8CbPtSvJgqqaPXy+f8lTkiQ1N56bPNeIqpo36BokSVJbjmBIkqTmDBiSJKk5A4YkSWrOgCFJkpozYEiSpOYMGJIkqTkDhiRJas6AIUmSmjNgSJKk5gwYkiSpOQOGJElqzoAhSZKaM2BIkqTmDBiSJKk5A4YkSWrOgCFJkpozYEiSpOYMGJIkqTkDhiRJas6AIUmSmps06AI2JDO33pILj5476DIkSRo4RzAkSVJzBgxJktScAUOSJDVnwJAkSc0ZMCRJUnMGDEmS1JwBQ5IkNWfAkCRJzRkwJElScwYMSZLUnAFDkiQ1Z8CQJEnNGTAkSVJzBgxJktScAUOSJDVnwJAkSc0ZMCRJUnMGDEmS1JwBQ5IkNWfAkCRJzRkwJElScwYMSZLUnAFDkiQ1Z8CQJEnNGTAkSVJzBgxJktRcqmrQNWwwkiwBfjToOjYg04H7B13EBsT+bMe+bMv+bGtt9+eOVbXt8JmT1mIBG4MfVdXsQRexoUgy3/5sx/5sx75sy/5sa13pTy+RSJKk5gwYkiSpOQNGW2cNuoANjP3Zlv3Zjn3Zlv3Z1jrRn97kKUmSmnMEQ5IkNWfAWA1JXpLkR0luS/KObt65Sa5L8oG+9d6V5PDBVbpuSXJHkuuTLEwyv5u3dZLLk9za/dyqm58kZ3R9fF2SZ3Xzd0myoJu3fzdvUpJvJpk2uKNb85J8KsnPktzQN29c/dctO7Zb/9Ykx3bzNk3yjSQ3JHlj37pn9W+7oRilL+clWdy9PhcmObhv2Tu7vvxRkhf3zf+D94Ju/kbzfpDkKUmuSHJTkhuTvKWb72tzNaykP9e/12dV+RjHA5gI3A7sDGwCLAL2BM7ull8OPB7YAfjaoOtdlx7AHcD0YfP+HnhHN/0O4EPd9MHAPwMBngN8t5v/YeBA4MnARd28NwPHDfr41kL//RHwLOCGx9B/WwM/7n5u1U1vBcwF3kXvPx3/3q27F3DOoI97LfblPODEEdbdvft3vimwU/fvf+Io7wW7b2zvB92xPaubfhxwS9cPvjbb9ud69/p0BGP89gVuq6ofV9XvgPOBlwJTk0wAJgMPA+8BTh5cmeuNw4DPdNOfAQ7vm//Z6vkOsGWSHYBlwLTusSzJlsChwGfXbtlrX1X9G/DLYbPH238vBi6vql9W1X/Se4N5Cb/v18n03vgB3gv8rzV1PIM0Sl+O5jDg/Kp6qKp+AtxG731gpPeCw+j15UbzflBVP62q73fTS4CbgSfha3O1rKQ/R7POvj4NGOP3JOCuvud3d/N+Dnwf+BrwNGDC0ItEjyjgsu4Sx/HdvO2q6qfd9L3Adt30aP18JvC39N6wPkDvTeYDVbViTRe/jhpv/402/3JgBvAd4Iwkc4HvV9U9a670ddKbuqHjTw0N6TPOvqyqm9lI3w+SzAD2Ab6Lr83HbFh/wnr2+vQveTZSVW8dmk7yNeD1SU6iN5R3eVV9cmDFrTsOrKrFSZ4AXJ7kh/0Lq6qSrPRjTVX1H8AcgCRPo3ep5OYkn6M3DPi/quqWNVL9Om4s/beSbZcDrwJIMhm4FDgsyYeBp9L7H+clzYpdN/1fev8zru7nacBrVqehjfH9IMnmwEXAW6vq10keWeZrc/xG6M/17vXpCMb4LQae0vf8yd08AJIcBiwANgdmVtWfAkdmA78BcSyqanH382fAV+gN4d3XDY/S/fxZt/pK+7nzfnrXZv8ncDbwN2ygw9ArMd7+G0u/vpHeJafnAL8C/gz46+aVr2Oq6r6qergbDfskvdcnPIa+3FjeD7oT/0XAuVX15W62r83VNFJ/ro+vTwPG+F0LPD3JTkk2AY4CLoFHXhRvpXdz01R6SRN6N9tsMoBa1xlJNkvyuKFp4EXADfT67thutWOBr3bTlwCv7u44fw7wq77hVpI8D7inqm6ld312RffYoN64x2C8/Xcp8KIkW3VDrC/q5gHQzTuE3pv4UL8WvdfzBm3oZNh5Gb3XJ/T68qju0ww7AU8HvsdK3gu69jaK94P0hirOAW6uqg/3LfK1uRpG68/18vXZ+q7RjeFB7y7oW+jdoXtS3/y30n2agd7NSOcB19PdPb0xP+jdybyoe9w41G/ANsC3gFuBbwJb9/XfmV0fXw/M7msr9K7LDq27G73ridcBBwz6WNdgH54H/JTeTVp3A69dzf57Db0bwW4D/nzYPv4BmNNNTwEu635fbx708a+Fvvxc11fX0Xsj3qFv/ZO6vvwR8Cd980d8L+iWbRTvB/Q+1VVdvy3sHgf72mzen+vd69O/5ClJkprzEokkSWrOgCFJkpozYEiSpOYMGJIkqTkDhiRJas6AIUmSmjNgSJKk5gwYkiSpuf8PePq28S+JtvQAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Plot configuration\n",
    "ax = _df.plot(kind = 'barh', color=['#53a4b1','#c34454'], figsize = (8,6))\n",
    "\n",
    "ax.xaxis.set_major_formatter(mtick.PercentFormatter())\n",
    "plt.legend(['Retain', 'Churn'],fancybox=True,shadow=True)\n",
    "plt.axes().get_yaxis().set_label_text('')\n",
    "plt.title('Phone Service Customer')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Internet Service Customer"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Task 9**:  \n",
    "Using the same steps as what you did on \"Phone Service Customer\" plot, except this time the plot should compare  the frequency proportion of churned and retained customers in each service of the internet subscription!"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Internet Service Customer')"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "## Your code here\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Customer Tenure\n",
    "\n",
    "The \"Customer Tenure\" tab contains explanatory analysis on how the tenure varies in churned and retained customers. There are two plots you need to recreate for this tab; **Churn Rate by Tenure Group** and **Average Lifetime Value by Tenure**\n",
    "\n",
    "### Churn Rate by Tenure Group\n",
    "\n",
    "**Task 10**:  \n",
    "The Churn Rate by Tenure Group plot compares the frequency proportion of churned and retained customers in each tenure group.\n",
    "\n",
    "You can start by create the frequency table for the visualization. Since we'll separate the bars by the `churn_label`, you'll have to create a data frame that has the `tenure_group` as row index and the values (Yes & No) of `churn label` as column."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "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>churn_label</th>\n",
       "      <th>No</th>\n",
       "      <th>Yes</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tenure_group</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>&lt; 1 Year</th>\n",
       "      <td>16.31</td>\n",
       "      <td>14.72</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1-2 Year</th>\n",
       "      <td>10.36</td>\n",
       "      <td>4.17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2-4 Year</th>\n",
       "      <td>18.02</td>\n",
       "      <td>4.61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4-5 Year</th>\n",
       "      <td>10.11</td>\n",
       "      <td>1.70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>&gt; 5 Year</th>\n",
       "      <td>18.66</td>\n",
       "      <td>1.32</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "churn_label      No    Yes\n",
       "tenure_group              \n",
       "< 1 Year      16.31  14.72\n",
       "1-2 Year      10.36   4.17\n",
       "2-4 Year      18.02   4.61\n",
       "4-5 Year      10.11   1.70\n",
       "> 5 Year      18.66   1.32"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Your code here\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Churn Rate by Tenure Group')"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = ______.plot(kind = 'bar', color=['#53a4b1','#c34454'], figsize=(8, 6))\n",
    "\n",
    "# Plot Configuration\n",
    "ax.yaxis.set_major_formatter(mtick.PercentFormatter())\n",
    "plt.axes().get_xaxis().set_label_text('')\n",
    "plt.xticks(rotation = 360)\n",
    "plt.legend(['Retain', 'Churn'],fancybox=True,shadow=True)\n",
    "plt.title('Churn Rate by Tenure Group')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Once your data is ready, recall how to use `.plot(kind = 'bar')` to create a bar plot. You can also enhance the visualization such as adjusting the color by add the color parameter to your plot function call `.plot(kind = 'bar',color=['#53a4b1','#c34454'])`.\n",
    "\n",
    "Again, **do not mind the plot's customization** as you can copy-and-paste the codes defined under the `plot_tenure_churn()` function in `helper.py`. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Average LTV by Tenure"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Task 11**:  \n",
    "The last plot you need to recreate is the Average Lifetime Value by Tenure plot. This plot visualize how the average customer life time value varies in each tenure month, both in retained and churned customers. \n",
    "\n",
    "Let's start by aggregating our data. Since we want to plot two lines; one line of churned customer and the other for the retained customer, we need to create a dataframe that has the `tenure_months` as row index and the average (`mean`) `cltv` values of both No & Yes `churn_label` in columns as below:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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 tr th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe thead tr:last-of-type th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr>\n",
       "      <th></th>\n",
       "      <th colspan=\"2\" halign=\"left\">cltv</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>churn_label</th>\n",
       "      <th>No</th>\n",
       "      <th>Yes</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>tenure_months</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>3665.545455</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3993.549356</td>\n",
       "      <td>4049.110526</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4300.539130</td>\n",
       "      <td>4013.398374</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>4042.216981</td>\n",
       "      <td>3915.904255</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3888.387097</td>\n",
       "      <td>4196.975904</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>68</th>\n",
       "      <td>5185.494505</td>\n",
       "      <td>5120.888889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>69</th>\n",
       "      <td>5187.862069</td>\n",
       "      <td>5133.375000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>70</th>\n",
       "      <td>5155.962963</td>\n",
       "      <td>5327.818182</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>71</th>\n",
       "      <td>5154.500000</td>\n",
       "      <td>4859.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>72</th>\n",
       "      <td>5248.632022</td>\n",
       "      <td>4975.166667</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>73 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                      cltv             \n",
       "churn_label             No          Yes\n",
       "tenure_months                          \n",
       "0              3665.545455          NaN\n",
       "1              3993.549356  4049.110526\n",
       "2              4300.539130  4013.398374\n",
       "3              4042.216981  3915.904255\n",
       "4              3888.387097  4196.975904\n",
       "...                    ...          ...\n",
       "68             5185.494505  5120.888889\n",
       "69             5187.862069  5133.375000\n",
       "70             5155.962963  5327.818182\n",
       "71             5154.500000  4859.000000\n",
       "72             5248.632022  4975.166667\n",
       "\n",
       "[73 rows x 2 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "## Your code here\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Once your data is ready, recall how to use `.plot(kind = 'line')` to create a line plot. You can also enhance the visualization such as adjusting the color by add the color parameter to your plot function call `.plot(kind = 'bar',color=['#333333','#b3b3b3'])`.\n",
    "\n",
    "Once again, **do not mind the plot's customization** as you can copy-and-paste the codes defined under the `plot_tenure_cltv()` function in `helper.py`. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Average Lifetime Value by Tenure')"
      ]
     },
     "execution_count": 51,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = ______.plot(color=['#333333','#b3b3b3'], figsize=(8, 6),style = '.--')\n",
    "\n",
    "ax.yaxis.set_major_formatter(mtick.StrMethodFormatter('${x:,.0f}'))\n",
    "plt.axes().get_xaxis().set_label_text('Tenure (in Months)')\n",
    "plt.xticks(rotation = 360)\n",
    "plt.legend(['Retain', 'Churn'],fancybox=True,shadow=True)\n",
    "plt.title('Average Lifetime Value by Tenure')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<span style=\"font-size:1.5em;font-weight:bold\">Update the application scripts!</span>\n",
    "\n",
    "Now that your plots are ready, it's time to move them to the dashboard!\n",
    "\n",
    "- **On helper.py**:\n",
    "    - Use the code on **Task 8-11** to complete the functions below:\n",
    "        - Task 8 --> `plot_phone()`\n",
    "        - Task 9 --> `plot_internet()`\n",
    "        - Task 10 --> `plot_tenure_churn()`\n",
    "        - Task 11 --> `plot_tenure_cltv()`\n",
    "        \n",
    "- **On app.py:**\n",
    "    - Now that all the functions are completed, you can call all functions from helper.py using the asterisk (\\*) symbol: `from helper import *` \n",
    "    - \"Uncomment\" the codes under `# generate plot` section to generate the plots\n",
    "    - Lastly, in `render_template()` at the end of `index()` in `app.py`, uncomment the remaining codes."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Create-Your-Own Analysis\n",
    "\n",
    "We have prepared a Create-Your-Own tab on the apps where you can play around with the data frame and come up with your own analysis perspective. Don't forget to do the following steps:\n",
    "\n",
    "- Pass a variable onto the `render_template()` from `index()`\n",
    "- Create a separate function on `helper.py` for data wrangling and preprocessing, either returning a dataframe, simple value, or a plot to keep the `app.py` as clean as possible and easy to debug at later times\n",
    "- Edit the `index.html` to make sure it display your intended output on the page. You can copy-and-paste the skeleton from the other plots or values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "## Your code here\n"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Capstone Metrasys",
   "language": "python",
   "name": "capstone"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.2"
  },
  "latex_envs": {
   "LaTeX_envs_menu_present": true,
   "autoclose": false,
   "autocomplete": true,
   "bibliofile": "biblio.bib",
   "cite_by": "apalike",
   "current_citInitial": 1,
   "eqLabelWithNumbers": true,
   "eqNumInitial": 1,
   "hotkeys": {
    "equation": "Ctrl-E",
    "itemize": "Ctrl-I"
   },
   "labels_anchors": false,
   "latex_user_defs": false,
   "report_style_numbering": false,
   "user_envs_cfg": false
  },
  "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": false,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "284.261px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}