eda_starter.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "a207638d-2d90-4a41-81e0-f86b35a174f9",
   "metadata": {},
   "source": [
    "# Exploratory Data Analysis Starter\n",
    "\n",
    "## Import packages"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "21fa22a3-62fc-4ee9-ad3b-5e3fbc059bdf",
   "metadata": {},
   "outputs": [],
   "source": [
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import pandas as pd\n",
    "\n",
    "# Shows plots in jupyter notebook\n",
    "%matplotlib inline\n",
    "\n",
    "# Set plot style\n",
    "sns.set(color_codes=True)\n",
    "\n",
    "import numpy as np\n",
    "\n",
    "import warnings\n",
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "plt.rcParams['figure.figsize']=15,8"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "88cb5312-cbc0-4193-b84e-2c94455c9edf",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "## Loading data with Pandas\n",
    "\n",
    "We need to load `client_data.csv` and `price_data.csv` into individual dataframes so that we can work with them in Python. For this notebook and all further notebooks, it will be assumed that the CSV files will the placed in the same file location as the notebook. If they are not, please adjust the directory within the `read_csv` method accordingly."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "82e6a4ae-2b8a-4152-b04f-0c998466fd50",
   "metadata": {},
   "outputs": [],
   "source": [
    "client_df = pd.read_csv('./client_data.csv')\n",
    "price_df = pd.read_csv('./price_data.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "4ac7aaab-7c26-40ca-8f63-cf2ef9541ce6",
   "metadata": {},
   "source": [
    "You can view the first 3 rows of a dataframe using the `head` method. Similarly, if you wanted to see the last 3, you can use `tail(3)`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "dcf0d488-d3df-4819-a52b-e335c0674e43",
   "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>id</th>\n",
       "      <th>channel_sales</th>\n",
       "      <th>cons_12m</th>\n",
       "      <th>cons_gas_12m</th>\n",
       "      <th>cons_last_month</th>\n",
       "      <th>date_activ</th>\n",
       "      <th>date_end</th>\n",
       "      <th>date_modif_prod</th>\n",
       "      <th>date_renewal</th>\n",
       "      <th>forecast_cons_12m</th>\n",
       "      <th>...</th>\n",
       "      <th>has_gas</th>\n",
       "      <th>imp_cons</th>\n",
       "      <th>margin_gross_pow_ele</th>\n",
       "      <th>margin_net_pow_ele</th>\n",
       "      <th>nb_prod_act</th>\n",
       "      <th>net_margin</th>\n",
       "      <th>num_years_antig</th>\n",
       "      <th>origin_up</th>\n",
       "      <th>pow_max</th>\n",
       "      <th>churn</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>24011ae4ebbe3035111d65fa7c15bc57</td>\n",
       "      <td>foosdfpfkusacimwkcsosbicdxkicaua</td>\n",
       "      <td>0</td>\n",
       "      <td>54946</td>\n",
       "      <td>0</td>\n",
       "      <td>2013-06-15</td>\n",
       "      <td>2016-06-15</td>\n",
       "      <td>2015-11-01</td>\n",
       "      <td>2015-06-23</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>t</td>\n",
       "      <td>0.0</td>\n",
       "      <td>25.44</td>\n",
       "      <td>25.44</td>\n",
       "      <td>2</td>\n",
       "      <td>678.99</td>\n",
       "      <td>3</td>\n",
       "      <td>lxidpiddsbxsbosboudacockeimpuepw</td>\n",
       "      <td>43.648</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>d29c2c54acc38ff3c0614d0a653813dd</td>\n",
       "      <td>MISSING</td>\n",
       "      <td>4660</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2009-08-21</td>\n",
       "      <td>2016-08-30</td>\n",
       "      <td>2009-08-21</td>\n",
       "      <td>2015-08-31</td>\n",
       "      <td>189.95</td>\n",
       "      <td>...</td>\n",
       "      <td>f</td>\n",
       "      <td>0.0</td>\n",
       "      <td>16.38</td>\n",
       "      <td>16.38</td>\n",
       "      <td>1</td>\n",
       "      <td>18.89</td>\n",
       "      <td>6</td>\n",
       "      <td>kamkkxfxxuwbdslkwifmmcsiusiuosws</td>\n",
       "      <td>13.800</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>764c75f661154dac3a6c254cd082ea7d</td>\n",
       "      <td>foosdfpfkusacimwkcsosbicdxkicaua</td>\n",
       "      <td>544</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2010-04-16</td>\n",
       "      <td>2016-04-16</td>\n",
       "      <td>2010-04-16</td>\n",
       "      <td>2015-04-17</td>\n",
       "      <td>47.96</td>\n",
       "      <td>...</td>\n",
       "      <td>f</td>\n",
       "      <td>0.0</td>\n",
       "      <td>28.60</td>\n",
       "      <td>28.60</td>\n",
       "      <td>1</td>\n",
       "      <td>6.60</td>\n",
       "      <td>6</td>\n",
       "      <td>kamkkxfxxuwbdslkwifmmcsiusiuosws</td>\n",
       "      <td>13.856</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 id                     channel_sales  \\\n",
       "0  24011ae4ebbe3035111d65fa7c15bc57  foosdfpfkusacimwkcsosbicdxkicaua   \n",
       "1  d29c2c54acc38ff3c0614d0a653813dd                           MISSING   \n",
       "2  764c75f661154dac3a6c254cd082ea7d  foosdfpfkusacimwkcsosbicdxkicaua   \n",
       "\n",
       "   cons_12m  cons_gas_12m  cons_last_month  date_activ    date_end  \\\n",
       "0         0         54946                0  2013-06-15  2016-06-15   \n",
       "1      4660             0                0  2009-08-21  2016-08-30   \n",
       "2       544             0                0  2010-04-16  2016-04-16   \n",
       "\n",
       "  date_modif_prod date_renewal  forecast_cons_12m  ...  has_gas  imp_cons  \\\n",
       "0      2015-11-01   2015-06-23               0.00  ...        t       0.0   \n",
       "1      2009-08-21   2015-08-31             189.95  ...        f       0.0   \n",
       "2      2010-04-16   2015-04-17              47.96  ...        f       0.0   \n",
       "\n",
       "   margin_gross_pow_ele  margin_net_pow_ele  nb_prod_act  net_margin  \\\n",
       "0                 25.44               25.44            2      678.99   \n",
       "1                 16.38               16.38            1       18.89   \n",
       "2                 28.60               28.60            1        6.60   \n",
       "\n",
       "  num_years_antig                         origin_up  pow_max  churn  \n",
       "0               3  lxidpiddsbxsbosboudacockeimpuepw   43.648      1  \n",
       "1               6  kamkkxfxxuwbdslkwifmmcsiusiuosws   13.800      0  \n",
       "2               6  kamkkxfxxuwbdslkwifmmcsiusiuosws   13.856      0  \n",
       "\n",
       "[3 rows x 26 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "client_df.head(3)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "2654a02d-8cf3-4a18-903e-b4f7f95d8e9e",
   "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>id</th>\n",
       "      <th>price_date</th>\n",
       "      <th>price_off_peak_var</th>\n",
       "      <th>price_peak_var</th>\n",
       "      <th>price_mid_peak_var</th>\n",
       "      <th>price_off_peak_fix</th>\n",
       "      <th>price_peak_fix</th>\n",
       "      <th>price_mid_peak_fix</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>2015-01-01</td>\n",
       "      <td>0.151367</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>44.266931</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>2015-02-01</td>\n",
       "      <td>0.151367</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>44.266931</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>2015-03-01</td>\n",
       "      <td>0.151367</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>44.266931</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 id  price_date  price_off_peak_var  \\\n",
       "0  038af19179925da21a25619c5a24b745  2015-01-01            0.151367   \n",
       "1  038af19179925da21a25619c5a24b745  2015-02-01            0.151367   \n",
       "2  038af19179925da21a25619c5a24b745  2015-03-01            0.151367   \n",
       "\n",
       "   price_peak_var  price_mid_peak_var  price_off_peak_fix  price_peak_fix  \\\n",
       "0             0.0                 0.0           44.266931             0.0   \n",
       "1             0.0                 0.0           44.266931             0.0   \n",
       "2             0.0                 0.0           44.266931             0.0   \n",
       "\n",
       "   price_mid_peak_fix  \n",
       "0                 0.0  \n",
       "1                 0.0  \n",
       "2                 0.0  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "price_df.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "966c3948-41c0-4183-9118-dd665c58a93c",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "## Descriptive statistics of data\n",
    "\n",
    "### Data types\n",
    "\n",
    "It is useful to first understand the data that you're dealing with along with the data types of each column. The data types may dictate how you transform and engineer features.\n",
    "\n",
    "To get an overview of the data types within a data frame, use the `info()` method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "2dc75b07-c286-4852-bc33-e501782e67d9",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 14606 entries, 0 to 14605\n",
      "Data columns (total 26 columns):\n",
      " #   Column                          Non-Null Count  Dtype  \n",
      "---  ------                          --------------  -----  \n",
      " 0   id                              14606 non-null  object \n",
      " 1   channel_sales                   14606 non-null  object \n",
      " 2   cons_12m                        14606 non-null  int64  \n",
      " 3   cons_gas_12m                    14606 non-null  int64  \n",
      " 4   cons_last_month                 14606 non-null  int64  \n",
      " 5   date_activ                      14606 non-null  object \n",
      " 6   date_end                        14606 non-null  object \n",
      " 7   date_modif_prod                 14606 non-null  object \n",
      " 8   date_renewal                    14606 non-null  object \n",
      " 9   forecast_cons_12m               14606 non-null  float64\n",
      " 10  forecast_cons_year              14606 non-null  int64  \n",
      " 11  forecast_discount_energy        14606 non-null  float64\n",
      " 12  forecast_meter_rent_12m         14606 non-null  float64\n",
      " 13  forecast_price_energy_off_peak  14606 non-null  float64\n",
      " 14  forecast_price_energy_peak      14606 non-null  float64\n",
      " 15  forecast_price_pow_off_peak     14606 non-null  float64\n",
      " 16  has_gas                         14606 non-null  object \n",
      " 17  imp_cons                        14606 non-null  float64\n",
      " 18  margin_gross_pow_ele            14606 non-null  float64\n",
      " 19  margin_net_pow_ele              14606 non-null  float64\n",
      " 20  nb_prod_act                     14606 non-null  int64  \n",
      " 21  net_margin                      14606 non-null  float64\n",
      " 22  num_years_antig                 14606 non-null  int64  \n",
      " 23  origin_up                       14606 non-null  object \n",
      " 24  pow_max                         14606 non-null  float64\n",
      " 25  churn                           14606 non-null  int64  \n",
      "dtypes: float64(11), int64(7), object(8)\n",
      "memory usage: 2.9+ MB\n"
     ]
    }
   ],
   "source": [
    "client_df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "b67f2628-351a-48ee-b8ba-88c54042af88",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 193002 entries, 0 to 193001\n",
      "Data columns (total 8 columns):\n",
      " #   Column              Non-Null Count   Dtype  \n",
      "---  ------              --------------   -----  \n",
      " 0   id                  193002 non-null  object \n",
      " 1   price_date          193002 non-null  object \n",
      " 2   price_off_peak_var  193002 non-null  float64\n",
      " 3   price_peak_var      193002 non-null  float64\n",
      " 4   price_mid_peak_var  193002 non-null  float64\n",
      " 5   price_off_peak_fix  193002 non-null  float64\n",
      " 6   price_peak_fix      193002 non-null  float64\n",
      " 7   price_mid_peak_fix  193002 non-null  float64\n",
      "dtypes: float64(6), object(2)\n",
      "memory usage: 11.8+ MB\n"
     ]
    }
   ],
   "source": [
    "price_df.info()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "3d7e80c5-ef07-47a7-908a-a081b3ebd09e",
   "metadata": {},
   "source": [
    "### Statistics\n",
    "\n",
    "Now let's look at some statistics about the datasets. We can do this by using the `describe()` method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "60f3010c-ecaf-4b8a-bb87-93bb54cc0c7e",
   "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>cons_12m</th>\n",
       "      <th>cons_gas_12m</th>\n",
       "      <th>cons_last_month</th>\n",
       "      <th>forecast_cons_12m</th>\n",
       "      <th>forecast_cons_year</th>\n",
       "      <th>forecast_discount_energy</th>\n",
       "      <th>forecast_meter_rent_12m</th>\n",
       "      <th>forecast_price_energy_off_peak</th>\n",
       "      <th>forecast_price_energy_peak</th>\n",
       "      <th>forecast_price_pow_off_peak</th>\n",
       "      <th>imp_cons</th>\n",
       "      <th>margin_gross_pow_ele</th>\n",
       "      <th>margin_net_pow_ele</th>\n",
       "      <th>nb_prod_act</th>\n",
       "      <th>net_margin</th>\n",
       "      <th>num_years_antig</th>\n",
       "      <th>pow_max</th>\n",
       "      <th>churn</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1.460600e+04</td>\n",
       "      <td>1.460600e+04</td>\n",
       "      <td>14606.000000</td>\n",
       "      <td>14606.000000</td>\n",
       "      <td>14606.000000</td>\n",
       "      <td>14606.000000</td>\n",
       "      <td>14606.000000</td>\n",
       "      <td>14606.000000</td>\n",
       "      <td>14606.000000</td>\n",
       "      <td>14606.000000</td>\n",
       "      <td>14606.000000</td>\n",
       "      <td>14606.000000</td>\n",
       "      <td>14606.000000</td>\n",
       "      <td>14606.000000</td>\n",
       "      <td>14606.000000</td>\n",
       "      <td>14606.000000</td>\n",
       "      <td>14606.000000</td>\n",
       "      <td>14606.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>1.592203e+05</td>\n",
       "      <td>2.809238e+04</td>\n",
       "      <td>16090.269752</td>\n",
       "      <td>1868.614880</td>\n",
       "      <td>1399.762906</td>\n",
       "      <td>0.966726</td>\n",
       "      <td>63.086871</td>\n",
       "      <td>0.137283</td>\n",
       "      <td>0.050491</td>\n",
       "      <td>43.130056</td>\n",
       "      <td>152.786896</td>\n",
       "      <td>24.565121</td>\n",
       "      <td>24.562517</td>\n",
       "      <td>1.292346</td>\n",
       "      <td>189.264522</td>\n",
       "      <td>4.997809</td>\n",
       "      <td>18.135136</td>\n",
       "      <td>0.097152</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>5.734653e+05</td>\n",
       "      <td>1.629731e+05</td>\n",
       "      <td>64364.196422</td>\n",
       "      <td>2387.571531</td>\n",
       "      <td>3247.786255</td>\n",
       "      <td>5.108289</td>\n",
       "      <td>66.165783</td>\n",
       "      <td>0.024623</td>\n",
       "      <td>0.049037</td>\n",
       "      <td>4.485988</td>\n",
       "      <td>341.369366</td>\n",
       "      <td>20.231172</td>\n",
       "      <td>20.230280</td>\n",
       "      <td>0.709774</td>\n",
       "      <td>311.798130</td>\n",
       "      <td>1.611749</td>\n",
       "      <td>13.534743</td>\n",
       "      <td>0.296175</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000e+00</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>0.000000</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>1.000000</td>\n",
       "      <td>3.300000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>5.674750e+03</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>494.995000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>16.180000</td>\n",
       "      <td>0.116340</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>40.606701</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>14.280000</td>\n",
       "      <td>14.280000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>50.712500</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>12.500000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>1.411550e+04</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>792.500000</td>\n",
       "      <td>1112.875000</td>\n",
       "      <td>314.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>18.795000</td>\n",
       "      <td>0.143166</td>\n",
       "      <td>0.084138</td>\n",
       "      <td>44.311378</td>\n",
       "      <td>37.395000</td>\n",
       "      <td>21.640000</td>\n",
       "      <td>21.640000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>112.530000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>13.856000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>4.076375e+04</td>\n",
       "      <td>0.000000e+00</td>\n",
       "      <td>3383.000000</td>\n",
       "      <td>2401.790000</td>\n",
       "      <td>1745.750000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>131.030000</td>\n",
       "      <td>0.146348</td>\n",
       "      <td>0.098837</td>\n",
       "      <td>44.311378</td>\n",
       "      <td>193.980000</td>\n",
       "      <td>29.880000</td>\n",
       "      <td>29.880000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>243.097500</td>\n",
       "      <td>6.000000</td>\n",
       "      <td>19.172500</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>6.207104e+06</td>\n",
       "      <td>4.154590e+06</td>\n",
       "      <td>771203.000000</td>\n",
       "      <td>82902.830000</td>\n",
       "      <td>175375.000000</td>\n",
       "      <td>30.000000</td>\n",
       "      <td>599.310000</td>\n",
       "      <td>0.273963</td>\n",
       "      <td>0.195975</td>\n",
       "      <td>59.266378</td>\n",
       "      <td>15042.790000</td>\n",
       "      <td>374.640000</td>\n",
       "      <td>374.640000</td>\n",
       "      <td>32.000000</td>\n",
       "      <td>24570.650000</td>\n",
       "      <td>13.000000</td>\n",
       "      <td>320.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           cons_12m  cons_gas_12m  cons_last_month  forecast_cons_12m  \\\n",
       "count  1.460600e+04  1.460600e+04     14606.000000       14606.000000   \n",
       "mean   1.592203e+05  2.809238e+04     16090.269752        1868.614880   \n",
       "std    5.734653e+05  1.629731e+05     64364.196422        2387.571531   \n",
       "min    0.000000e+00  0.000000e+00         0.000000           0.000000   \n",
       "25%    5.674750e+03  0.000000e+00         0.000000         494.995000   \n",
       "50%    1.411550e+04  0.000000e+00       792.500000        1112.875000   \n",
       "75%    4.076375e+04  0.000000e+00      3383.000000        2401.790000   \n",
       "max    6.207104e+06  4.154590e+06    771203.000000       82902.830000   \n",
       "\n",
       "       forecast_cons_year  forecast_discount_energy  forecast_meter_rent_12m  \\\n",
       "count        14606.000000              14606.000000             14606.000000   \n",
       "mean          1399.762906                  0.966726                63.086871   \n",
       "std           3247.786255                  5.108289                66.165783   \n",
       "min              0.000000                  0.000000                 0.000000   \n",
       "25%              0.000000                  0.000000                16.180000   \n",
       "50%            314.000000                  0.000000                18.795000   \n",
       "75%           1745.750000                  0.000000               131.030000   \n",
       "max         175375.000000                 30.000000               599.310000   \n",
       "\n",
       "       forecast_price_energy_off_peak  forecast_price_energy_peak  \\\n",
       "count                    14606.000000                14606.000000   \n",
       "mean                         0.137283                    0.050491   \n",
       "std                          0.024623                    0.049037   \n",
       "min                          0.000000                    0.000000   \n",
       "25%                          0.116340                    0.000000   \n",
       "50%                          0.143166                    0.084138   \n",
       "75%                          0.146348                    0.098837   \n",
       "max                          0.273963                    0.195975   \n",
       "\n",
       "       forecast_price_pow_off_peak      imp_cons  margin_gross_pow_ele  \\\n",
       "count                 14606.000000  14606.000000          14606.000000   \n",
       "mean                     43.130056    152.786896             24.565121   \n",
       "std                       4.485988    341.369366             20.231172   \n",
       "min                       0.000000      0.000000              0.000000   \n",
       "25%                      40.606701      0.000000             14.280000   \n",
       "50%                      44.311378     37.395000             21.640000   \n",
       "75%                      44.311378    193.980000             29.880000   \n",
       "max                      59.266378  15042.790000            374.640000   \n",
       "\n",
       "       margin_net_pow_ele   nb_prod_act    net_margin  num_years_antig  \\\n",
       "count        14606.000000  14606.000000  14606.000000     14606.000000   \n",
       "mean            24.562517      1.292346    189.264522         4.997809   \n",
       "std             20.230280      0.709774    311.798130         1.611749   \n",
       "min              0.000000      1.000000      0.000000         1.000000   \n",
       "25%             14.280000      1.000000     50.712500         4.000000   \n",
       "50%             21.640000      1.000000    112.530000         5.000000   \n",
       "75%             29.880000      1.000000    243.097500         6.000000   \n",
       "max            374.640000     32.000000  24570.650000        13.000000   \n",
       "\n",
       "            pow_max         churn  \n",
       "count  14606.000000  14606.000000  \n",
       "mean      18.135136      0.097152  \n",
       "std       13.534743      0.296175  \n",
       "min        3.300000      0.000000  \n",
       "25%       12.500000      0.000000  \n",
       "50%       13.856000      0.000000  \n",
       "75%       19.172500      0.000000  \n",
       "max      320.000000      1.000000  "
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "client_df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "f8b442e6-e186-42f9-b3c5-c5c15cd8c82a",
   "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>price_off_peak_var</th>\n",
       "      <th>price_peak_var</th>\n",
       "      <th>price_mid_peak_var</th>\n",
       "      <th>price_off_peak_fix</th>\n",
       "      <th>price_peak_fix</th>\n",
       "      <th>price_mid_peak_fix</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>193002.000000</td>\n",
       "      <td>193002.000000</td>\n",
       "      <td>193002.000000</td>\n",
       "      <td>193002.000000</td>\n",
       "      <td>193002.000000</td>\n",
       "      <td>193002.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>0.141027</td>\n",
       "      <td>0.054630</td>\n",
       "      <td>0.030496</td>\n",
       "      <td>43.334477</td>\n",
       "      <td>10.622875</td>\n",
       "      <td>6.409984</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>0.025032</td>\n",
       "      <td>0.049924</td>\n",
       "      <td>0.036298</td>\n",
       "      <td>5.410297</td>\n",
       "      <td>12.841895</td>\n",
       "      <td>7.773592</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>0.125976</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>40.728885</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>0.146033</td>\n",
       "      <td>0.085483</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>44.266930</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>0.151635</td>\n",
       "      <td>0.101673</td>\n",
       "      <td>0.072558</td>\n",
       "      <td>44.444710</td>\n",
       "      <td>24.339581</td>\n",
       "      <td>16.226389</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>0.280700</td>\n",
       "      <td>0.229788</td>\n",
       "      <td>0.114102</td>\n",
       "      <td>59.444710</td>\n",
       "      <td>36.490692</td>\n",
       "      <td>17.458221</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       price_off_peak_var  price_peak_var  price_mid_peak_var  \\\n",
       "count       193002.000000   193002.000000       193002.000000   \n",
       "mean             0.141027        0.054630            0.030496   \n",
       "std              0.025032        0.049924            0.036298   \n",
       "min              0.000000        0.000000            0.000000   \n",
       "25%              0.125976        0.000000            0.000000   \n",
       "50%              0.146033        0.085483            0.000000   \n",
       "75%              0.151635        0.101673            0.072558   \n",
       "max              0.280700        0.229788            0.114102   \n",
       "\n",
       "       price_off_peak_fix  price_peak_fix  price_mid_peak_fix  \n",
       "count       193002.000000   193002.000000       193002.000000  \n",
       "mean            43.334477       10.622875            6.409984  \n",
       "std              5.410297       12.841895            7.773592  \n",
       "min              0.000000        0.000000            0.000000  \n",
       "25%             40.728885        0.000000            0.000000  \n",
       "50%             44.266930        0.000000            0.000000  \n",
       "75%             44.444710       24.339581           16.226389  \n",
       "max             59.444710       36.490692           17.458221  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "price_df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6a7a8879-c483-4464-9057-c70e6b150e53",
   "metadata": {},
   "source": [
    "---\n",
    "\n",
    "## Data visualization\n",
    "\n",
    "If you're working in Python, two of the most popular packages for visualization are `matplotlib` and `seaborn`. We highly recommend you use these, or at least be familiar with them because they are ubiquitous!\n",
    "\n",
    "Below are some functions that you can use to get started with visualizations. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "f350ea0a-dc45-43f5-a6b7-e5e816964419",
   "metadata": {},
   "outputs": [],
   "source": [
    "def plot_stacked_bars(dataframe, title_, size_=(18, 10), rot_=0, legend_=\"upper right\"):\n",
    "    \"\"\"\n",
    "    Plot stacked bars with annotations\n",
    "    \"\"\"\n",
    "    ax = dataframe.plot(\n",
    "        kind=\"bar\",\n",
    "        stacked=True,\n",
    "        figsize=size_,\n",
    "        rot=rot_,\n",
    "        title=title_\n",
    "    )\n",
    "\n",
    "    # Annotate bars\n",
    "    annotate_stacked_bars(ax, textsize=14)\n",
    "    # Rename legend\n",
    "    plt.legend([\"Retention\", \"Churn\"], loc=legend_)\n",
    "    # Labels\n",
    "    plt.ylabel(\"Company base (%)\")\n",
    "    plt.show()\n",
    "\n",
    "def annotate_stacked_bars(ax, pad=0.99, colour=\"white\", textsize=13):\n",
    "    \"\"\"\n",
    "    Add value annotations to the bars\n",
    "    \"\"\"\n",
    "\n",
    "    # Iterate over the plotted rectanges/bars\n",
    "    for p in ax.patches:\n",
    "        \n",
    "        # Calculate annotation\n",
    "        value = str(round(p.get_height(),1))\n",
    "        # If value is 0 do not annotate\n",
    "        if value == '0.0':\n",
    "            continue\n",
    "        ax.annotate(\n",
    "            value,\n",
    "            ((p.get_x()+ p.get_width()/2)*pad-0.05, (p.get_y()+p.get_height()/2)*pad),\n",
    "            color=colour,\n",
    "            size=textsize\n",
    "        )\n",
    "\n",
    "def plot_distribution(dataframe, column, ax, bins_=50):\n",
    "    \"\"\"\n",
    "    Plot variable distirbution in a stacked histogram of churned or retained company\n",
    "    \"\"\"\n",
    "    # Create a temporal dataframe with the data to be plot\n",
    "    temp = pd.DataFrame({\"Retention\": dataframe[dataframe[\"churn\"]==0][column],\n",
    "    \"Churn\":dataframe[dataframe[\"churn\"]==1][column]})\n",
    "    # Plot the histogram\n",
    "    temp[[\"Retention\",\"Churn\"]].plot(kind='hist', bins=bins_, ax=ax, stacked=True)\n",
    "    # X-axis label\n",
    "    ax.set_xlabel(column)\n",
    "    # Change the x-axis to plain style\n",
    "    ax.ticklabel_format(style='plain', axis='x')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6eab14d0",
   "metadata": {},
   "source": [
    "Thhe first function `plot_stacked_bars` is used to plot a stacked bar chart. An example of how you could use this is shown below:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "448c05d8",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "churn = client_df[['id', 'churn']]\n",
    "churn.columns = ['Companies', 'churn']\n",
    "churn_total = churn.groupby(churn['churn']).count()\n",
    "\n",
    "churn_percentage = churn_total / churn_total.sum() * 100\n",
    "plot_stacked_bars(churn_percentage.transpose(), \"Churning status\", (5, 5), legend_=\"lower right\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d055f10e",
   "metadata": {},
   "source": [
    "The second function `annotate_bars` is used by the first function, but the third function `plot_distribution` helps you to plot the distribution of a numeric column. An example of how it can be used is given below:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "9a28757e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "consumption = client_df[['id', 'cons_12m', 'cons_gas_12m', 'cons_last_month', 'imp_cons', 'has_gas', 'churn']]\n",
    "\n",
    "fig, axs = plt.subplots(nrows=1, figsize=(18, 5))\n",
    "\n",
    "plot_distribution(consumption, 'cons_12m', axs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "57a3ee64",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x360 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "for i in consumption.columns[1:5]:\n",
    "    fig, axs = plt.subplots(nrows=1, figsize=(18, 5))\n",
    "    plot_distribution(consumption, i, axs)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "ebaa7798",
   "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>id</th>\n",
       "      <th>channel_sales</th>\n",
       "      <th>cons_12m</th>\n",
       "      <th>cons_gas_12m</th>\n",
       "      <th>cons_last_month</th>\n",
       "      <th>date_activ</th>\n",
       "      <th>date_end</th>\n",
       "      <th>date_modif_prod</th>\n",
       "      <th>date_renewal</th>\n",
       "      <th>forecast_cons_12m</th>\n",
       "      <th>...</th>\n",
       "      <th>has_gas</th>\n",
       "      <th>imp_cons</th>\n",
       "      <th>margin_gross_pow_ele</th>\n",
       "      <th>margin_net_pow_ele</th>\n",
       "      <th>nb_prod_act</th>\n",
       "      <th>net_margin</th>\n",
       "      <th>num_years_antig</th>\n",
       "      <th>origin_up</th>\n",
       "      <th>pow_max</th>\n",
       "      <th>churn</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>24011ae4ebbe3035111d65fa7c15bc57</td>\n",
       "      <td>foosdfpfkusacimwkcsosbicdxkicaua</td>\n",
       "      <td>0</td>\n",
       "      <td>54946</td>\n",
       "      <td>0</td>\n",
       "      <td>2013-06-15</td>\n",
       "      <td>2016-06-15</td>\n",
       "      <td>2015-11-01</td>\n",
       "      <td>2015-06-23</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>t</td>\n",
       "      <td>0.00</td>\n",
       "      <td>25.44</td>\n",
       "      <td>25.44</td>\n",
       "      <td>2</td>\n",
       "      <td>678.99</td>\n",
       "      <td>3</td>\n",
       "      <td>lxidpiddsbxsbosboudacockeimpuepw</td>\n",
       "      <td>43.648</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>d29c2c54acc38ff3c0614d0a653813dd</td>\n",
       "      <td>MISSING</td>\n",
       "      <td>4660</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2009-08-21</td>\n",
       "      <td>2016-08-30</td>\n",
       "      <td>2009-08-21</td>\n",
       "      <td>2015-08-31</td>\n",
       "      <td>189.95</td>\n",
       "      <td>...</td>\n",
       "      <td>f</td>\n",
       "      <td>0.00</td>\n",
       "      <td>16.38</td>\n",
       "      <td>16.38</td>\n",
       "      <td>1</td>\n",
       "      <td>18.89</td>\n",
       "      <td>6</td>\n",
       "      <td>kamkkxfxxuwbdslkwifmmcsiusiuosws</td>\n",
       "      <td>13.800</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>764c75f661154dac3a6c254cd082ea7d</td>\n",
       "      <td>foosdfpfkusacimwkcsosbicdxkicaua</td>\n",
       "      <td>544</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2010-04-16</td>\n",
       "      <td>2016-04-16</td>\n",
       "      <td>2010-04-16</td>\n",
       "      <td>2015-04-17</td>\n",
       "      <td>47.96</td>\n",
       "      <td>...</td>\n",
       "      <td>f</td>\n",
       "      <td>0.00</td>\n",
       "      <td>28.60</td>\n",
       "      <td>28.60</td>\n",
       "      <td>1</td>\n",
       "      <td>6.60</td>\n",
       "      <td>6</td>\n",
       "      <td>kamkkxfxxuwbdslkwifmmcsiusiuosws</td>\n",
       "      <td>13.856</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>bba03439a292a1e166f80264c16191cb</td>\n",
       "      <td>lmkebamcaaclubfxadlmueccxoimlema</td>\n",
       "      <td>1584</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2010-03-30</td>\n",
       "      <td>2016-03-30</td>\n",
       "      <td>2010-03-30</td>\n",
       "      <td>2015-03-31</td>\n",
       "      <td>240.04</td>\n",
       "      <td>...</td>\n",
       "      <td>f</td>\n",
       "      <td>0.00</td>\n",
       "      <td>30.22</td>\n",
       "      <td>30.22</td>\n",
       "      <td>1</td>\n",
       "      <td>25.46</td>\n",
       "      <td>6</td>\n",
       "      <td>kamkkxfxxuwbdslkwifmmcsiusiuosws</td>\n",
       "      <td>13.200</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>149d57cf92fc41cf94415803a877cb4b</td>\n",
       "      <td>MISSING</td>\n",
       "      <td>4425</td>\n",
       "      <td>0</td>\n",
       "      <td>526</td>\n",
       "      <td>2010-01-13</td>\n",
       "      <td>2016-03-07</td>\n",
       "      <td>2010-01-13</td>\n",
       "      <td>2015-03-09</td>\n",
       "      <td>445.75</td>\n",
       "      <td>...</td>\n",
       "      <td>f</td>\n",
       "      <td>52.32</td>\n",
       "      <td>44.91</td>\n",
       "      <td>44.91</td>\n",
       "      <td>1</td>\n",
       "      <td>47.98</td>\n",
       "      <td>6</td>\n",
       "      <td>kamkkxfxxuwbdslkwifmmcsiusiuosws</td>\n",
       "      <td>19.800</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 26 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 id                     channel_sales  \\\n",
       "0  24011ae4ebbe3035111d65fa7c15bc57  foosdfpfkusacimwkcsosbicdxkicaua   \n",
       "1  d29c2c54acc38ff3c0614d0a653813dd                           MISSING   \n",
       "2  764c75f661154dac3a6c254cd082ea7d  foosdfpfkusacimwkcsosbicdxkicaua   \n",
       "3  bba03439a292a1e166f80264c16191cb  lmkebamcaaclubfxadlmueccxoimlema   \n",
       "4  149d57cf92fc41cf94415803a877cb4b                           MISSING   \n",
       "\n",
       "   cons_12m  cons_gas_12m  cons_last_month  date_activ    date_end  \\\n",
       "0         0         54946                0  2013-06-15  2016-06-15   \n",
       "1      4660             0                0  2009-08-21  2016-08-30   \n",
       "2       544             0                0  2010-04-16  2016-04-16   \n",
       "3      1584             0                0  2010-03-30  2016-03-30   \n",
       "4      4425             0              526  2010-01-13  2016-03-07   \n",
       "\n",
       "  date_modif_prod date_renewal  forecast_cons_12m  ...  has_gas  imp_cons  \\\n",
       "0      2015-11-01   2015-06-23               0.00  ...        t      0.00   \n",
       "1      2009-08-21   2015-08-31             189.95  ...        f      0.00   \n",
       "2      2010-04-16   2015-04-17              47.96  ...        f      0.00   \n",
       "3      2010-03-30   2015-03-31             240.04  ...        f      0.00   \n",
       "4      2010-01-13   2015-03-09             445.75  ...        f     52.32   \n",
       "\n",
       "   margin_gross_pow_ele  margin_net_pow_ele  nb_prod_act  net_margin  \\\n",
       "0                 25.44               25.44            2      678.99   \n",
       "1                 16.38               16.38            1       18.89   \n",
       "2                 28.60               28.60            1        6.60   \n",
       "3                 30.22               30.22            1       25.46   \n",
       "4                 44.91               44.91            1       47.98   \n",
       "\n",
       "  num_years_antig                         origin_up  pow_max  churn  \n",
       "0               3  lxidpiddsbxsbosboudacockeimpuepw   43.648      1  \n",
       "1               6  kamkkxfxxuwbdslkwifmmcsiusiuosws   13.800      0  \n",
       "2               6  kamkkxfxxuwbdslkwifmmcsiusiuosws   13.856      0  \n",
       "3               6  kamkkxfxxuwbdslkwifmmcsiusiuosws   13.200      0  \n",
       "4               6  kamkkxfxxuwbdslkwifmmcsiusiuosws   19.800      0  \n",
       "\n",
       "[5 rows x 26 columns]"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "client_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "1617a70a",
   "metadata": {},
   "outputs": [],
   "source": [
    "dfc=client_df.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "1a052ee9",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Missing values \n",
    "#dfc.isnull().sum()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "f0eee95a",
   "metadata": {},
   "outputs": [],
   "source": [
    "### Missing value is absent but they are stored as string in the data.Lets do a value counts to know such things.\n",
    "### Certain columns require datatype change before proceeding further"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "4d85d4e8",
   "metadata": {},
   "outputs": [],
   "source": [
    "dfc.iloc[:,5:9]=dfc.iloc[:,5:9].astype(\"datetime64[ns]\")  #DATETIME object wrongly stored as string\n",
    "dfc['churn']=dfc['churn'].astype('object')                #categorical column given as int."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "22c3b009",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Segregate dataframe to respective datatype\n",
    "dfc_cat=dfc.select_dtypes('object')\n",
    "dfc_num=dfc.select_dtypes('number')\n",
    "dfc_date=dfc.select_dtypes('datetime')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "2c21f55e",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "foosdfpfkusacimwkcsosbicdxkicaua    6754\n",
      "MISSING                             3725\n",
      "lmkebamcaaclubfxadlmueccxoimlema    1843\n",
      "usilxuppasemubllopkaafesmlibmsdf    1375\n",
      "ewpakwlliwisiwduibdlfmalxowmwpci     893\n",
      "sddiedcslfslkckwlfkdpoeeailfpeds      11\n",
      "epumfxlbckeskwekxbiuasklxalciiuu       3\n",
      "fixdbufsefwooaasfcxdxadsiekoceaa       2\n",
      "Name: channel_sales, dtype: int64\n",
      "\n",
      "\n",
      "\n",
      "f    11955\n",
      "t     2651\n",
      "Name: has_gas, dtype: int64\n",
      "\n",
      "\n",
      "\n",
      "lxidpiddsbxsbosboudacockeimpuepw    7097\n",
      "kamkkxfxxuwbdslkwifmmcsiusiuosws    4294\n",
      "ldkssxwpmemidmecebumciepifcamkci    3148\n",
      "MISSING                               64\n",
      "usapbepcfoloekilkwsdiboslwaxobdp       2\n",
      "ewxeelcelemmiwuafmddpobolfuxioce       1\n",
      "Name: origin_up, dtype: int64\n",
      "\n",
      "\n",
      "\n",
      "0    13187\n",
      "1     1419\n",
      "Name: churn, dtype: int64\n",
      "\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Lets check for value_counts for cat columns\n",
    "\n",
    "for i in dfc_cat.columns[1:]:\n",
    "    print(dfc_cat.loc[:,i].value_counts())\n",
    "    print('\\n\\n')\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "ca00afc3",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "id                  0\n",
       "channel_sales    3725\n",
       "has_gas             0\n",
       "origin_up          64\n",
       "churn               0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "### MISSING values should be replaced with np.nan\n",
    "for i in dfc_cat.columns[1:]:\n",
    "    dfc_cat.loc[:,i].replace({'MISSING':np.nan},inplace=True)\n",
    "    \n",
    "    \n",
    "dfc_cat.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "id": "411961d4",
   "metadata": {},
   "outputs": [],
   "source": [
    "yt=dfc.churn.astype('int')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "id": "7b381c06",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(14606, 17)"
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfc_num.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "id": "eafa15eb",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 2160x2160 with 18 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "f,ax=plt.subplots(nrows=6, ncols=3,figsize=(30,30))\n",
    "\n",
    "for variable, subplot in zip(dfc_num.columns,ax.flatten()):\n",
    "    sns.boxplot(x=yt,y=dfc_num[variable],ax=subplot)\n",
    "    \n",
    "plt.tight_layout()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "id": "780db828",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Only outcome of this plot 'num_years_antig' has a little effect on deciding churn."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 93,
   "id": "d6a11c62",
   "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>cons_12m</th>\n",
       "      <th>cons_gas_12m</th>\n",
       "      <th>cons_last_month</th>\n",
       "      <th>forecast_cons_12m</th>\n",
       "      <th>forecast_cons_year</th>\n",
       "      <th>forecast_discount_energy</th>\n",
       "      <th>forecast_meter_rent_12m</th>\n",
       "      <th>forecast_price_energy_off_peak</th>\n",
       "      <th>forecast_price_energy_peak</th>\n",
       "      <th>forecast_price_pow_off_peak</th>\n",
       "      <th>imp_cons</th>\n",
       "      <th>margin_gross_pow_ele</th>\n",
       "      <th>margin_net_pow_ele</th>\n",
       "      <th>nb_prod_act</th>\n",
       "      <th>net_margin</th>\n",
       "      <th>num_years_antig</th>\n",
       "      <th>pow_max</th>\n",
       "      <th>churn</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>cons_12m</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.488474</td>\n",
       "      <td>0.968212</td>\n",
       "      <td>0.193947</td>\n",
       "      <td>0.167093</td>\n",
       "      <td>-0.043282</td>\n",
       "      <td>0.065268</td>\n",
       "      <td>-0.007748</td>\n",
       "      <td>0.145908</td>\n",
       "      <td>-0.026566</td>\n",
       "      <td>0.159711</td>\n",
       "      <td>-0.011926</td>\n",
       "      <td>-0.011905</td>\n",
       "      <td>0.154251</td>\n",
       "      <td>0.133614</td>\n",
       "      <td>-0.003565</td>\n",
       "      <td>0.082889</td>\n",
       "      <td>-0.045968</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cons_gas_12m</th>\n",
       "      <td>0.488474</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.507007</td>\n",
       "      <td>0.084359</td>\n",
       "      <td>0.080934</td>\n",
       "      <td>-0.012595</td>\n",
       "      <td>0.041393</td>\n",
       "      <td>-0.017684</td>\n",
       "      <td>0.074002</td>\n",
       "      <td>-0.020558</td>\n",
       "      <td>0.077846</td>\n",
       "      <td>0.006868</td>\n",
       "      <td>0.006852</td>\n",
       "      <td>0.239387</td>\n",
       "      <td>0.070867</td>\n",
       "      <td>-0.013815</td>\n",
       "      <td>0.054317</td>\n",
       "      <td>-0.037957</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cons_last_month</th>\n",
       "      <td>0.968212</td>\n",
       "      <td>0.507007</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.177773</td>\n",
       "      <td>0.193574</td>\n",
       "      <td>-0.040874</td>\n",
       "      <td>0.057476</td>\n",
       "      <td>-0.005187</td>\n",
       "      <td>0.136802</td>\n",
       "      <td>-0.023017</td>\n",
       "      <td>0.187034</td>\n",
       "      <td>-0.011498</td>\n",
       "      <td>-0.011477</td>\n",
       "      <td>0.169099</td>\n",
       "      <td>0.121835</td>\n",
       "      <td>-0.003677</td>\n",
       "      <td>0.074529</td>\n",
       "      <td>-0.045284</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>forecast_cons_12m</th>\n",
       "      <td>0.193947</td>\n",
       "      <td>0.084359</td>\n",
       "      <td>0.177773</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.647727</td>\n",
       "      <td>0.058435</td>\n",
       "      <td>0.305627</td>\n",
       "      <td>-0.135646</td>\n",
       "      <td>0.254056</td>\n",
       "      <td>-0.018477</td>\n",
       "      <td>0.634600</td>\n",
       "      <td>-0.023220</td>\n",
       "      <td>-0.023231</td>\n",
       "      <td>0.055915</td>\n",
       "      <td>0.768537</td>\n",
       "      <td>0.021283</td>\n",
       "      <td>0.393946</td>\n",
       "      <td>0.012949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>forecast_cons_year</th>\n",
       "      <td>0.167093</td>\n",
       "      <td>0.080934</td>\n",
       "      <td>0.193574</td>\n",
       "      <td>0.647727</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.008518</td>\n",
       "      <td>0.276009</td>\n",
       "      <td>-0.158012</td>\n",
       "      <td>0.251005</td>\n",
       "      <td>-0.044190</td>\n",
       "      <td>0.969395</td>\n",
       "      <td>-0.010759</td>\n",
       "      <td>-0.010727</td>\n",
       "      <td>0.040426</td>\n",
       "      <td>0.464928</td>\n",
       "      <td>0.048248</td>\n",
       "      <td>0.309647</td>\n",
       "      <td>-0.002558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>forecast_discount_energy</th>\n",
       "      <td>-0.043282</td>\n",
       "      <td>-0.012595</td>\n",
       "      <td>-0.040874</td>\n",
       "      <td>0.058435</td>\n",
       "      <td>0.008518</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.008388</td>\n",
       "      <td>0.353735</td>\n",
       "      <td>0.059318</td>\n",
       "      <td>0.048024</td>\n",
       "      <td>0.037939</td>\n",
       "      <td>0.241142</td>\n",
       "      <td>0.240421</td>\n",
       "      <td>0.106812</td>\n",
       "      <td>0.079784</td>\n",
       "      <td>-0.070697</td>\n",
       "      <td>-0.005032</td>\n",
       "      <td>0.017026</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>forecast_meter_rent_12m</th>\n",
       "      <td>0.065268</td>\n",
       "      <td>0.041393</td>\n",
       "      <td>0.057476</td>\n",
       "      <td>0.305627</td>\n",
       "      <td>0.276009</td>\n",
       "      <td>-0.008388</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.579353</td>\n",
       "      <td>0.706376</td>\n",
       "      <td>-0.203089</td>\n",
       "      <td>0.224212</td>\n",
       "      <td>0.183509</td>\n",
       "      <td>0.183566</td>\n",
       "      <td>0.018109</td>\n",
       "      <td>0.275830</td>\n",
       "      <td>0.071558</td>\n",
       "      <td>0.618549</td>\n",
       "      <td>0.044245</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>forecast_price_energy_off_peak</th>\n",
       "      <td>-0.007748</td>\n",
       "      <td>-0.017684</td>\n",
       "      <td>-0.005187</td>\n",
       "      <td>-0.135646</td>\n",
       "      <td>-0.158012</td>\n",
       "      <td>0.353735</td>\n",
       "      <td>-0.579353</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.330138</td>\n",
       "      <td>0.630377</td>\n",
       "      <td>-0.092535</td>\n",
       "      <td>0.094498</td>\n",
       "      <td>0.094216</td>\n",
       "      <td>0.035093</td>\n",
       "      <td>-0.135107</td>\n",
       "      <td>-0.168095</td>\n",
       "      <td>-0.384397</td>\n",
       "      <td>-0.010837</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>forecast_price_energy_peak</th>\n",
       "      <td>0.145908</td>\n",
       "      <td>0.074002</td>\n",
       "      <td>0.136802</td>\n",
       "      <td>0.254056</td>\n",
       "      <td>0.251005</td>\n",
       "      <td>0.059318</td>\n",
       "      <td>0.706376</td>\n",
       "      <td>-0.330138</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.242017</td>\n",
       "      <td>0.212317</td>\n",
       "      <td>0.165483</td>\n",
       "      <td>0.165560</td>\n",
       "      <td>0.044141</td>\n",
       "      <td>0.234942</td>\n",
       "      <td>0.080260</td>\n",
       "      <td>0.411532</td>\n",
       "      <td>0.029315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>forecast_price_pow_off_peak</th>\n",
       "      <td>-0.026566</td>\n",
       "      <td>-0.020558</td>\n",
       "      <td>-0.023017</td>\n",
       "      <td>-0.018477</td>\n",
       "      <td>-0.044190</td>\n",
       "      <td>0.048024</td>\n",
       "      <td>-0.203089</td>\n",
       "      <td>0.630377</td>\n",
       "      <td>-0.242017</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.011318</td>\n",
       "      <td>-0.061869</td>\n",
       "      <td>-0.061925</td>\n",
       "      <td>-0.005335</td>\n",
       "      <td>-0.072623</td>\n",
       "      <td>-0.120136</td>\n",
       "      <td>-0.095897</td>\n",
       "      <td>0.014778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>imp_cons</th>\n",
       "      <td>0.159711</td>\n",
       "      <td>0.077846</td>\n",
       "      <td>0.187034</td>\n",
       "      <td>0.634600</td>\n",
       "      <td>0.969395</td>\n",
       "      <td>0.037939</td>\n",
       "      <td>0.224212</td>\n",
       "      <td>-0.092535</td>\n",
       "      <td>0.212317</td>\n",
       "      <td>-0.011318</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.002099</td>\n",
       "      <td>-0.002084</td>\n",
       "      <td>0.046166</td>\n",
       "      <td>0.458502</td>\n",
       "      <td>0.024947</td>\n",
       "      <td>0.260808</td>\n",
       "      <td>-0.001583</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>margin_gross_pow_ele</th>\n",
       "      <td>-0.011926</td>\n",
       "      <td>0.006868</td>\n",
       "      <td>-0.011498</td>\n",
       "      <td>-0.023220</td>\n",
       "      <td>-0.010759</td>\n",
       "      <td>0.241142</td>\n",
       "      <td>0.183509</td>\n",
       "      <td>0.094498</td>\n",
       "      <td>0.165483</td>\n",
       "      <td>-0.061869</td>\n",
       "      <td>-0.002099</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.999914</td>\n",
       "      <td>-0.006784</td>\n",
       "      <td>0.031814</td>\n",
       "      <td>-0.061091</td>\n",
       "      <td>0.373298</td>\n",
       "      <td>0.095725</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>margin_net_pow_ele</th>\n",
       "      <td>-0.011905</td>\n",
       "      <td>0.006852</td>\n",
       "      <td>-0.011477</td>\n",
       "      <td>-0.023231</td>\n",
       "      <td>-0.010727</td>\n",
       "      <td>0.240421</td>\n",
       "      <td>0.183566</td>\n",
       "      <td>0.094216</td>\n",
       "      <td>0.165560</td>\n",
       "      <td>-0.061925</td>\n",
       "      <td>-0.002084</td>\n",
       "      <td>0.999914</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.006763</td>\n",
       "      <td>0.031639</td>\n",
       "      <td>-0.061197</td>\n",
       "      <td>0.373351</td>\n",
       "      <td>0.095772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>nb_prod_act</th>\n",
       "      <td>0.154251</td>\n",
       "      <td>0.239387</td>\n",
       "      <td>0.169099</td>\n",
       "      <td>0.055915</td>\n",
       "      <td>0.040426</td>\n",
       "      <td>0.106812</td>\n",
       "      <td>0.018109</td>\n",
       "      <td>0.035093</td>\n",
       "      <td>0.044141</td>\n",
       "      <td>-0.005335</td>\n",
       "      <td>0.046166</td>\n",
       "      <td>-0.006784</td>\n",
       "      <td>-0.006763</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.053781</td>\n",
       "      <td>0.005228</td>\n",
       "      <td>0.012712</td>\n",
       "      <td>-0.014930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>net_margin</th>\n",
       "      <td>0.133614</td>\n",
       "      <td>0.070867</td>\n",
       "      <td>0.121835</td>\n",
       "      <td>0.768537</td>\n",
       "      <td>0.464928</td>\n",
       "      <td>0.079784</td>\n",
       "      <td>0.275830</td>\n",
       "      <td>-0.135107</td>\n",
       "      <td>0.234942</td>\n",
       "      <td>-0.072623</td>\n",
       "      <td>0.458502</td>\n",
       "      <td>0.031814</td>\n",
       "      <td>0.031639</td>\n",
       "      <td>0.053781</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.000263</td>\n",
       "      <td>0.325417</td>\n",
       "      <td>0.041135</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>num_years_antig</th>\n",
       "      <td>-0.003565</td>\n",
       "      <td>-0.013815</td>\n",
       "      <td>-0.003677</td>\n",
       "      <td>0.021283</td>\n",
       "      <td>0.048248</td>\n",
       "      <td>-0.070697</td>\n",
       "      <td>0.071558</td>\n",
       "      <td>-0.168095</td>\n",
       "      <td>0.080260</td>\n",
       "      <td>-0.120136</td>\n",
       "      <td>0.024947</td>\n",
       "      <td>-0.061091</td>\n",
       "      <td>-0.061197</td>\n",
       "      <td>0.005228</td>\n",
       "      <td>-0.000263</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.059715</td>\n",
       "      <td>-0.074140</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pow_max</th>\n",
       "      <td>0.082889</td>\n",
       "      <td>0.054317</td>\n",
       "      <td>0.074529</td>\n",
       "      <td>0.393946</td>\n",
       "      <td>0.309647</td>\n",
       "      <td>-0.005032</td>\n",
       "      <td>0.618549</td>\n",
       "      <td>-0.384397</td>\n",
       "      <td>0.411532</td>\n",
       "      <td>-0.095897</td>\n",
       "      <td>0.260808</td>\n",
       "      <td>0.373298</td>\n",
       "      <td>0.373351</td>\n",
       "      <td>0.012712</td>\n",
       "      <td>0.325417</td>\n",
       "      <td>0.059715</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.030362</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>churn</th>\n",
       "      <td>-0.045968</td>\n",
       "      <td>-0.037957</td>\n",
       "      <td>-0.045284</td>\n",
       "      <td>0.012949</td>\n",
       "      <td>-0.002558</td>\n",
       "      <td>0.017026</td>\n",
       "      <td>0.044245</td>\n",
       "      <td>-0.010837</td>\n",
       "      <td>0.029315</td>\n",
       "      <td>0.014778</td>\n",
       "      <td>-0.001583</td>\n",
       "      <td>0.095725</td>\n",
       "      <td>0.095772</td>\n",
       "      <td>-0.014930</td>\n",
       "      <td>0.041135</td>\n",
       "      <td>-0.074140</td>\n",
       "      <td>0.030362</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                cons_12m  cons_gas_12m  cons_last_month  \\\n",
       "cons_12m                        1.000000      0.488474         0.968212   \n",
       "cons_gas_12m                    0.488474      1.000000         0.507007   \n",
       "cons_last_month                 0.968212      0.507007         1.000000   \n",
       "forecast_cons_12m               0.193947      0.084359         0.177773   \n",
       "forecast_cons_year              0.167093      0.080934         0.193574   \n",
       "forecast_discount_energy       -0.043282     -0.012595        -0.040874   \n",
       "forecast_meter_rent_12m         0.065268      0.041393         0.057476   \n",
       "forecast_price_energy_off_peak -0.007748     -0.017684        -0.005187   \n",
       "forecast_price_energy_peak      0.145908      0.074002         0.136802   \n",
       "forecast_price_pow_off_peak    -0.026566     -0.020558        -0.023017   \n",
       "imp_cons                        0.159711      0.077846         0.187034   \n",
       "margin_gross_pow_ele           -0.011926      0.006868        -0.011498   \n",
       "margin_net_pow_ele             -0.011905      0.006852        -0.011477   \n",
       "nb_prod_act                     0.154251      0.239387         0.169099   \n",
       "net_margin                      0.133614      0.070867         0.121835   \n",
       "num_years_antig                -0.003565     -0.013815        -0.003677   \n",
       "pow_max                         0.082889      0.054317         0.074529   \n",
       "churn                          -0.045968     -0.037957        -0.045284   \n",
       "\n",
       "                                forecast_cons_12m  forecast_cons_year  \\\n",
       "cons_12m                                 0.193947            0.167093   \n",
       "cons_gas_12m                             0.084359            0.080934   \n",
       "cons_last_month                          0.177773            0.193574   \n",
       "forecast_cons_12m                        1.000000            0.647727   \n",
       "forecast_cons_year                       0.647727            1.000000   \n",
       "forecast_discount_energy                 0.058435            0.008518   \n",
       "forecast_meter_rent_12m                  0.305627            0.276009   \n",
       "forecast_price_energy_off_peak          -0.135646           -0.158012   \n",
       "forecast_price_energy_peak               0.254056            0.251005   \n",
       "forecast_price_pow_off_peak             -0.018477           -0.044190   \n",
       "imp_cons                                 0.634600            0.969395   \n",
       "margin_gross_pow_ele                    -0.023220           -0.010759   \n",
       "margin_net_pow_ele                      -0.023231           -0.010727   \n",
       "nb_prod_act                              0.055915            0.040426   \n",
       "net_margin                               0.768537            0.464928   \n",
       "num_years_antig                          0.021283            0.048248   \n",
       "pow_max                                  0.393946            0.309647   \n",
       "churn                                    0.012949           -0.002558   \n",
       "\n",
       "                                forecast_discount_energy  \\\n",
       "cons_12m                                       -0.043282   \n",
       "cons_gas_12m                                   -0.012595   \n",
       "cons_last_month                                -0.040874   \n",
       "forecast_cons_12m                               0.058435   \n",
       "forecast_cons_year                              0.008518   \n",
       "forecast_discount_energy                        1.000000   \n",
       "forecast_meter_rent_12m                        -0.008388   \n",
       "forecast_price_energy_off_peak                  0.353735   \n",
       "forecast_price_energy_peak                      0.059318   \n",
       "forecast_price_pow_off_peak                     0.048024   \n",
       "imp_cons                                        0.037939   \n",
       "margin_gross_pow_ele                            0.241142   \n",
       "margin_net_pow_ele                              0.240421   \n",
       "nb_prod_act                                     0.106812   \n",
       "net_margin                                      0.079784   \n",
       "num_years_antig                                -0.070697   \n",
       "pow_max                                        -0.005032   \n",
       "churn                                           0.017026   \n",
       "\n",
       "                                forecast_meter_rent_12m  \\\n",
       "cons_12m                                       0.065268   \n",
       "cons_gas_12m                                   0.041393   \n",
       "cons_last_month                                0.057476   \n",
       "forecast_cons_12m                              0.305627   \n",
       "forecast_cons_year                             0.276009   \n",
       "forecast_discount_energy                      -0.008388   \n",
       "forecast_meter_rent_12m                        1.000000   \n",
       "forecast_price_energy_off_peak                -0.579353   \n",
       "forecast_price_energy_peak                     0.706376   \n",
       "forecast_price_pow_off_peak                   -0.203089   \n",
       "imp_cons                                       0.224212   \n",
       "margin_gross_pow_ele                           0.183509   \n",
       "margin_net_pow_ele                             0.183566   \n",
       "nb_prod_act                                    0.018109   \n",
       "net_margin                                     0.275830   \n",
       "num_years_antig                                0.071558   \n",
       "pow_max                                        0.618549   \n",
       "churn                                          0.044245   \n",
       "\n",
       "                                forecast_price_energy_off_peak  \\\n",
       "cons_12m                                             -0.007748   \n",
       "cons_gas_12m                                         -0.017684   \n",
       "cons_last_month                                      -0.005187   \n",
       "forecast_cons_12m                                    -0.135646   \n",
       "forecast_cons_year                                   -0.158012   \n",
       "forecast_discount_energy                              0.353735   \n",
       "forecast_meter_rent_12m                              -0.579353   \n",
       "forecast_price_energy_off_peak                        1.000000   \n",
       "forecast_price_energy_peak                           -0.330138   \n",
       "forecast_price_pow_off_peak                           0.630377   \n",
       "imp_cons                                             -0.092535   \n",
       "margin_gross_pow_ele                                  0.094498   \n",
       "margin_net_pow_ele                                    0.094216   \n",
       "nb_prod_act                                           0.035093   \n",
       "net_margin                                           -0.135107   \n",
       "num_years_antig                                      -0.168095   \n",
       "pow_max                                              -0.384397   \n",
       "churn                                                -0.010837   \n",
       "\n",
       "                                forecast_price_energy_peak  \\\n",
       "cons_12m                                          0.145908   \n",
       "cons_gas_12m                                      0.074002   \n",
       "cons_last_month                                   0.136802   \n",
       "forecast_cons_12m                                 0.254056   \n",
       "forecast_cons_year                                0.251005   \n",
       "forecast_discount_energy                          0.059318   \n",
       "forecast_meter_rent_12m                           0.706376   \n",
       "forecast_price_energy_off_peak                   -0.330138   \n",
       "forecast_price_energy_peak                        1.000000   \n",
       "forecast_price_pow_off_peak                      -0.242017   \n",
       "imp_cons                                          0.212317   \n",
       "margin_gross_pow_ele                              0.165483   \n",
       "margin_net_pow_ele                                0.165560   \n",
       "nb_prod_act                                       0.044141   \n",
       "net_margin                                        0.234942   \n",
       "num_years_antig                                   0.080260   \n",
       "pow_max                                           0.411532   \n",
       "churn                                             0.029315   \n",
       "\n",
       "                                forecast_price_pow_off_peak  imp_cons  \\\n",
       "cons_12m                                          -0.026566  0.159711   \n",
       "cons_gas_12m                                      -0.020558  0.077846   \n",
       "cons_last_month                                   -0.023017  0.187034   \n",
       "forecast_cons_12m                                 -0.018477  0.634600   \n",
       "forecast_cons_year                                -0.044190  0.969395   \n",
       "forecast_discount_energy                           0.048024  0.037939   \n",
       "forecast_meter_rent_12m                           -0.203089  0.224212   \n",
       "forecast_price_energy_off_peak                     0.630377 -0.092535   \n",
       "forecast_price_energy_peak                        -0.242017  0.212317   \n",
       "forecast_price_pow_off_peak                        1.000000 -0.011318   \n",
       "imp_cons                                          -0.011318  1.000000   \n",
       "margin_gross_pow_ele                              -0.061869 -0.002099   \n",
       "margin_net_pow_ele                                -0.061925 -0.002084   \n",
       "nb_prod_act                                       -0.005335  0.046166   \n",
       "net_margin                                        -0.072623  0.458502   \n",
       "num_years_antig                                   -0.120136  0.024947   \n",
       "pow_max                                           -0.095897  0.260808   \n",
       "churn                                              0.014778 -0.001583   \n",
       "\n",
       "                                margin_gross_pow_ele  margin_net_pow_ele  \\\n",
       "cons_12m                                   -0.011926           -0.011905   \n",
       "cons_gas_12m                                0.006868            0.006852   \n",
       "cons_last_month                            -0.011498           -0.011477   \n",
       "forecast_cons_12m                          -0.023220           -0.023231   \n",
       "forecast_cons_year                         -0.010759           -0.010727   \n",
       "forecast_discount_energy                    0.241142            0.240421   \n",
       "forecast_meter_rent_12m                     0.183509            0.183566   \n",
       "forecast_price_energy_off_peak              0.094498            0.094216   \n",
       "forecast_price_energy_peak                  0.165483            0.165560   \n",
       "forecast_price_pow_off_peak                -0.061869           -0.061925   \n",
       "imp_cons                                   -0.002099           -0.002084   \n",
       "margin_gross_pow_ele                        1.000000            0.999914   \n",
       "margin_net_pow_ele                          0.999914            1.000000   \n",
       "nb_prod_act                                -0.006784           -0.006763   \n",
       "net_margin                                  0.031814            0.031639   \n",
       "num_years_antig                            -0.061091           -0.061197   \n",
       "pow_max                                     0.373298            0.373351   \n",
       "churn                                       0.095725            0.095772   \n",
       "\n",
       "                                nb_prod_act  net_margin  num_years_antig  \\\n",
       "cons_12m                           0.154251    0.133614        -0.003565   \n",
       "cons_gas_12m                       0.239387    0.070867        -0.013815   \n",
       "cons_last_month                    0.169099    0.121835        -0.003677   \n",
       "forecast_cons_12m                  0.055915    0.768537         0.021283   \n",
       "forecast_cons_year                 0.040426    0.464928         0.048248   \n",
       "forecast_discount_energy           0.106812    0.079784        -0.070697   \n",
       "forecast_meter_rent_12m            0.018109    0.275830         0.071558   \n",
       "forecast_price_energy_off_peak     0.035093   -0.135107        -0.168095   \n",
       "forecast_price_energy_peak         0.044141    0.234942         0.080260   \n",
       "forecast_price_pow_off_peak       -0.005335   -0.072623        -0.120136   \n",
       "imp_cons                           0.046166    0.458502         0.024947   \n",
       "margin_gross_pow_ele              -0.006784    0.031814        -0.061091   \n",
       "margin_net_pow_ele                -0.006763    0.031639        -0.061197   \n",
       "nb_prod_act                        1.000000    0.053781         0.005228   \n",
       "net_margin                         0.053781    1.000000        -0.000263   \n",
       "num_years_antig                    0.005228   -0.000263         1.000000   \n",
       "pow_max                            0.012712    0.325417         0.059715   \n",
       "churn                             -0.014930    0.041135        -0.074140   \n",
       "\n",
       "                                 pow_max     churn  \n",
       "cons_12m                        0.082889 -0.045968  \n",
       "cons_gas_12m                    0.054317 -0.037957  \n",
       "cons_last_month                 0.074529 -0.045284  \n",
       "forecast_cons_12m               0.393946  0.012949  \n",
       "forecast_cons_year              0.309647 -0.002558  \n",
       "forecast_discount_energy       -0.005032  0.017026  \n",
       "forecast_meter_rent_12m         0.618549  0.044245  \n",
       "forecast_price_energy_off_peak -0.384397 -0.010837  \n",
       "forecast_price_energy_peak      0.411532  0.029315  \n",
       "forecast_price_pow_off_peak    -0.095897  0.014778  \n",
       "imp_cons                        0.260808 -0.001583  \n",
       "margin_gross_pow_ele            0.373298  0.095725  \n",
       "margin_net_pow_ele              0.373351  0.095772  \n",
       "nb_prod_act                     0.012712 -0.014930  \n",
       "net_margin                      0.325417  0.041135  \n",
       "num_years_antig                 0.059715 -0.074140  \n",
       "pow_max                         1.000000  0.030362  \n",
       "churn                           0.030362  1.000000  "
      ]
     },
     "execution_count": 93,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfc_num=pd.concat([dfc_num,y],1)\n",
    "dfc_num.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 94,
   "id": "ea3c0e84",
   "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>cons_12m</th>\n",
       "      <th>cons_gas_12m</th>\n",
       "      <th>cons_last_month</th>\n",
       "      <th>forecast_cons_12m</th>\n",
       "      <th>forecast_cons_year</th>\n",
       "      <th>forecast_discount_energy</th>\n",
       "      <th>forecast_meter_rent_12m</th>\n",
       "      <th>forecast_price_energy_off_peak</th>\n",
       "      <th>forecast_price_energy_peak</th>\n",
       "      <th>forecast_price_pow_off_peak</th>\n",
       "      <th>imp_cons</th>\n",
       "      <th>margin_gross_pow_ele</th>\n",
       "      <th>margin_net_pow_ele</th>\n",
       "      <th>nb_prod_act</th>\n",
       "      <th>net_margin</th>\n",
       "      <th>num_years_antig</th>\n",
       "      <th>pow_max</th>\n",
       "      <th>churn</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>cons_12m</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.488474</td>\n",
       "      <td>0.968212</td>\n",
       "      <td>0.193947</td>\n",
       "      <td>0.167093</td>\n",
       "      <td>-0.043282</td>\n",
       "      <td>0.065268</td>\n",
       "      <td>-0.007748</td>\n",
       "      <td>0.145908</td>\n",
       "      <td>-0.026566</td>\n",
       "      <td>0.159711</td>\n",
       "      <td>-0.011926</td>\n",
       "      <td>-0.011905</td>\n",
       "      <td>0.154251</td>\n",
       "      <td>0.133614</td>\n",
       "      <td>-0.003565</td>\n",
       "      <td>0.082889</td>\n",
       "      <td>-0.045968</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cons_gas_12m</th>\n",
       "      <td>0.488474</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.507007</td>\n",
       "      <td>0.084359</td>\n",
       "      <td>0.080934</td>\n",
       "      <td>-0.012595</td>\n",
       "      <td>0.041393</td>\n",
       "      <td>-0.017684</td>\n",
       "      <td>0.074002</td>\n",
       "      <td>-0.020558</td>\n",
       "      <td>0.077846</td>\n",
       "      <td>0.006868</td>\n",
       "      <td>0.006852</td>\n",
       "      <td>0.239387</td>\n",
       "      <td>0.070867</td>\n",
       "      <td>-0.013815</td>\n",
       "      <td>0.054317</td>\n",
       "      <td>-0.037957</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>cons_last_month</th>\n",
       "      <td>0.968212</td>\n",
       "      <td>0.507007</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.177773</td>\n",
       "      <td>0.193574</td>\n",
       "      <td>-0.040874</td>\n",
       "      <td>0.057476</td>\n",
       "      <td>-0.005187</td>\n",
       "      <td>0.136802</td>\n",
       "      <td>-0.023017</td>\n",
       "      <td>0.187034</td>\n",
       "      <td>-0.011498</td>\n",
       "      <td>-0.011477</td>\n",
       "      <td>0.169099</td>\n",
       "      <td>0.121835</td>\n",
       "      <td>-0.003677</td>\n",
       "      <td>0.074529</td>\n",
       "      <td>-0.045284</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>forecast_cons_12m</th>\n",
       "      <td>0.193947</td>\n",
       "      <td>0.084359</td>\n",
       "      <td>0.177773</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.647727</td>\n",
       "      <td>0.058435</td>\n",
       "      <td>0.305627</td>\n",
       "      <td>-0.135646</td>\n",
       "      <td>0.254056</td>\n",
       "      <td>-0.018477</td>\n",
       "      <td>0.634600</td>\n",
       "      <td>-0.023220</td>\n",
       "      <td>-0.023231</td>\n",
       "      <td>0.055915</td>\n",
       "      <td>0.768537</td>\n",
       "      <td>0.021283</td>\n",
       "      <td>0.393946</td>\n",
       "      <td>0.012949</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>forecast_cons_year</th>\n",
       "      <td>0.167093</td>\n",
       "      <td>0.080934</td>\n",
       "      <td>0.193574</td>\n",
       "      <td>0.647727</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.008518</td>\n",
       "      <td>0.276009</td>\n",
       "      <td>-0.158012</td>\n",
       "      <td>0.251005</td>\n",
       "      <td>-0.044190</td>\n",
       "      <td>0.969395</td>\n",
       "      <td>-0.010759</td>\n",
       "      <td>-0.010727</td>\n",
       "      <td>0.040426</td>\n",
       "      <td>0.464928</td>\n",
       "      <td>0.048248</td>\n",
       "      <td>0.309647</td>\n",
       "      <td>-0.002558</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>forecast_discount_energy</th>\n",
       "      <td>-0.043282</td>\n",
       "      <td>-0.012595</td>\n",
       "      <td>-0.040874</td>\n",
       "      <td>0.058435</td>\n",
       "      <td>0.008518</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.008388</td>\n",
       "      <td>0.353735</td>\n",
       "      <td>0.059318</td>\n",
       "      <td>0.048024</td>\n",
       "      <td>0.037939</td>\n",
       "      <td>0.241142</td>\n",
       "      <td>0.240421</td>\n",
       "      <td>0.106812</td>\n",
       "      <td>0.079784</td>\n",
       "      <td>-0.070697</td>\n",
       "      <td>-0.005032</td>\n",
       "      <td>0.017026</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>forecast_meter_rent_12m</th>\n",
       "      <td>0.065268</td>\n",
       "      <td>0.041393</td>\n",
       "      <td>0.057476</td>\n",
       "      <td>0.305627</td>\n",
       "      <td>0.276009</td>\n",
       "      <td>-0.008388</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.579353</td>\n",
       "      <td>0.706376</td>\n",
       "      <td>-0.203089</td>\n",
       "      <td>0.224212</td>\n",
       "      <td>0.183509</td>\n",
       "      <td>0.183566</td>\n",
       "      <td>0.018109</td>\n",
       "      <td>0.275830</td>\n",
       "      <td>0.071558</td>\n",
       "      <td>0.618549</td>\n",
       "      <td>0.044245</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>forecast_price_energy_off_peak</th>\n",
       "      <td>-0.007748</td>\n",
       "      <td>-0.017684</td>\n",
       "      <td>-0.005187</td>\n",
       "      <td>-0.135646</td>\n",
       "      <td>-0.158012</td>\n",
       "      <td>0.353735</td>\n",
       "      <td>-0.579353</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.330138</td>\n",
       "      <td>0.630377</td>\n",
       "      <td>-0.092535</td>\n",
       "      <td>0.094498</td>\n",
       "      <td>0.094216</td>\n",
       "      <td>0.035093</td>\n",
       "      <td>-0.135107</td>\n",
       "      <td>-0.168095</td>\n",
       "      <td>-0.384397</td>\n",
       "      <td>-0.010837</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>forecast_price_energy_peak</th>\n",
       "      <td>0.145908</td>\n",
       "      <td>0.074002</td>\n",
       "      <td>0.136802</td>\n",
       "      <td>0.254056</td>\n",
       "      <td>0.251005</td>\n",
       "      <td>0.059318</td>\n",
       "      <td>0.706376</td>\n",
       "      <td>-0.330138</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.242017</td>\n",
       "      <td>0.212317</td>\n",
       "      <td>0.165483</td>\n",
       "      <td>0.165560</td>\n",
       "      <td>0.044141</td>\n",
       "      <td>0.234942</td>\n",
       "      <td>0.080260</td>\n",
       "      <td>0.411532</td>\n",
       "      <td>0.029315</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>forecast_price_pow_off_peak</th>\n",
       "      <td>-0.026566</td>\n",
       "      <td>-0.020558</td>\n",
       "      <td>-0.023017</td>\n",
       "      <td>-0.018477</td>\n",
       "      <td>-0.044190</td>\n",
       "      <td>0.048024</td>\n",
       "      <td>-0.203089</td>\n",
       "      <td>0.630377</td>\n",
       "      <td>-0.242017</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.011318</td>\n",
       "      <td>-0.061869</td>\n",
       "      <td>-0.061925</td>\n",
       "      <td>-0.005335</td>\n",
       "      <td>-0.072623</td>\n",
       "      <td>-0.120136</td>\n",
       "      <td>-0.095897</td>\n",
       "      <td>0.014778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>imp_cons</th>\n",
       "      <td>0.159711</td>\n",
       "      <td>0.077846</td>\n",
       "      <td>0.187034</td>\n",
       "      <td>0.634600</td>\n",
       "      <td>0.969395</td>\n",
       "      <td>0.037939</td>\n",
       "      <td>0.224212</td>\n",
       "      <td>-0.092535</td>\n",
       "      <td>0.212317</td>\n",
       "      <td>-0.011318</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.002099</td>\n",
       "      <td>-0.002084</td>\n",
       "      <td>0.046166</td>\n",
       "      <td>0.458502</td>\n",
       "      <td>0.024947</td>\n",
       "      <td>0.260808</td>\n",
       "      <td>-0.001583</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>margin_gross_pow_ele</th>\n",
       "      <td>-0.011926</td>\n",
       "      <td>0.006868</td>\n",
       "      <td>-0.011498</td>\n",
       "      <td>-0.023220</td>\n",
       "      <td>-0.010759</td>\n",
       "      <td>0.241142</td>\n",
       "      <td>0.183509</td>\n",
       "      <td>0.094498</td>\n",
       "      <td>0.165483</td>\n",
       "      <td>-0.061869</td>\n",
       "      <td>-0.002099</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.999914</td>\n",
       "      <td>-0.006784</td>\n",
       "      <td>0.031814</td>\n",
       "      <td>-0.061091</td>\n",
       "      <td>0.373298</td>\n",
       "      <td>0.095725</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>margin_net_pow_ele</th>\n",
       "      <td>-0.011905</td>\n",
       "      <td>0.006852</td>\n",
       "      <td>-0.011477</td>\n",
       "      <td>-0.023231</td>\n",
       "      <td>-0.010727</td>\n",
       "      <td>0.240421</td>\n",
       "      <td>0.183566</td>\n",
       "      <td>0.094216</td>\n",
       "      <td>0.165560</td>\n",
       "      <td>-0.061925</td>\n",
       "      <td>-0.002084</td>\n",
       "      <td>0.999914</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.006763</td>\n",
       "      <td>0.031639</td>\n",
       "      <td>-0.061197</td>\n",
       "      <td>0.373351</td>\n",
       "      <td>0.095772</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>nb_prod_act</th>\n",
       "      <td>0.154251</td>\n",
       "      <td>0.239387</td>\n",
       "      <td>0.169099</td>\n",
       "      <td>0.055915</td>\n",
       "      <td>0.040426</td>\n",
       "      <td>0.106812</td>\n",
       "      <td>0.018109</td>\n",
       "      <td>0.035093</td>\n",
       "      <td>0.044141</td>\n",
       "      <td>-0.005335</td>\n",
       "      <td>0.046166</td>\n",
       "      <td>-0.006784</td>\n",
       "      <td>-0.006763</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.053781</td>\n",
       "      <td>0.005228</td>\n",
       "      <td>0.012712</td>\n",
       "      <td>-0.014930</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>net_margin</th>\n",
       "      <td>0.133614</td>\n",
       "      <td>0.070867</td>\n",
       "      <td>0.121835</td>\n",
       "      <td>0.768537</td>\n",
       "      <td>0.464928</td>\n",
       "      <td>0.079784</td>\n",
       "      <td>0.275830</td>\n",
       "      <td>-0.135107</td>\n",
       "      <td>0.234942</td>\n",
       "      <td>-0.072623</td>\n",
       "      <td>0.458502</td>\n",
       "      <td>0.031814</td>\n",
       "      <td>0.031639</td>\n",
       "      <td>0.053781</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.000263</td>\n",
       "      <td>0.325417</td>\n",
       "      <td>0.041135</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>num_years_antig</th>\n",
       "      <td>-0.003565</td>\n",
       "      <td>-0.013815</td>\n",
       "      <td>-0.003677</td>\n",
       "      <td>0.021283</td>\n",
       "      <td>0.048248</td>\n",
       "      <td>-0.070697</td>\n",
       "      <td>0.071558</td>\n",
       "      <td>-0.168095</td>\n",
       "      <td>0.080260</td>\n",
       "      <td>-0.120136</td>\n",
       "      <td>0.024947</td>\n",
       "      <td>-0.061091</td>\n",
       "      <td>-0.061197</td>\n",
       "      <td>0.005228</td>\n",
       "      <td>-0.000263</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.059715</td>\n",
       "      <td>-0.074140</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>pow_max</th>\n",
       "      <td>0.082889</td>\n",
       "      <td>0.054317</td>\n",
       "      <td>0.074529</td>\n",
       "      <td>0.393946</td>\n",
       "      <td>0.309647</td>\n",
       "      <td>-0.005032</td>\n",
       "      <td>0.618549</td>\n",
       "      <td>-0.384397</td>\n",
       "      <td>0.411532</td>\n",
       "      <td>-0.095897</td>\n",
       "      <td>0.260808</td>\n",
       "      <td>0.373298</td>\n",
       "      <td>0.373351</td>\n",
       "      <td>0.012712</td>\n",
       "      <td>0.325417</td>\n",
       "      <td>0.059715</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.030362</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>churn</th>\n",
       "      <td>-0.045968</td>\n",
       "      <td>-0.037957</td>\n",
       "      <td>-0.045284</td>\n",
       "      <td>0.012949</td>\n",
       "      <td>-0.002558</td>\n",
       "      <td>0.017026</td>\n",
       "      <td>0.044245</td>\n",
       "      <td>-0.010837</td>\n",
       "      <td>0.029315</td>\n",
       "      <td>0.014778</td>\n",
       "      <td>-0.001583</td>\n",
       "      <td>0.095725</td>\n",
       "      <td>0.095772</td>\n",
       "      <td>-0.014930</td>\n",
       "      <td>0.041135</td>\n",
       "      <td>-0.074140</td>\n",
       "      <td>0.030362</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                cons_12m  cons_gas_12m  cons_last_month  \\\n",
       "cons_12m                        1.000000      0.488474         0.968212   \n",
       "cons_gas_12m                    0.488474      1.000000         0.507007   \n",
       "cons_last_month                 0.968212      0.507007         1.000000   \n",
       "forecast_cons_12m               0.193947      0.084359         0.177773   \n",
       "forecast_cons_year              0.167093      0.080934         0.193574   \n",
       "forecast_discount_energy       -0.043282     -0.012595        -0.040874   \n",
       "forecast_meter_rent_12m         0.065268      0.041393         0.057476   \n",
       "forecast_price_energy_off_peak -0.007748     -0.017684        -0.005187   \n",
       "forecast_price_energy_peak      0.145908      0.074002         0.136802   \n",
       "forecast_price_pow_off_peak    -0.026566     -0.020558        -0.023017   \n",
       "imp_cons                        0.159711      0.077846         0.187034   \n",
       "margin_gross_pow_ele           -0.011926      0.006868        -0.011498   \n",
       "margin_net_pow_ele             -0.011905      0.006852        -0.011477   \n",
       "nb_prod_act                     0.154251      0.239387         0.169099   \n",
       "net_margin                      0.133614      0.070867         0.121835   \n",
       "num_years_antig                -0.003565     -0.013815        -0.003677   \n",
       "pow_max                         0.082889      0.054317         0.074529   \n",
       "churn                          -0.045968     -0.037957        -0.045284   \n",
       "\n",
       "                                forecast_cons_12m  forecast_cons_year  \\\n",
       "cons_12m                                 0.193947            0.167093   \n",
       "cons_gas_12m                             0.084359            0.080934   \n",
       "cons_last_month                          0.177773            0.193574   \n",
       "forecast_cons_12m                        1.000000            0.647727   \n",
       "forecast_cons_year                       0.647727            1.000000   \n",
       "forecast_discount_energy                 0.058435            0.008518   \n",
       "forecast_meter_rent_12m                  0.305627            0.276009   \n",
       "forecast_price_energy_off_peak          -0.135646           -0.158012   \n",
       "forecast_price_energy_peak               0.254056            0.251005   \n",
       "forecast_price_pow_off_peak             -0.018477           -0.044190   \n",
       "imp_cons                                 0.634600            0.969395   \n",
       "margin_gross_pow_ele                    -0.023220           -0.010759   \n",
       "margin_net_pow_ele                      -0.023231           -0.010727   \n",
       "nb_prod_act                              0.055915            0.040426   \n",
       "net_margin                               0.768537            0.464928   \n",
       "num_years_antig                          0.021283            0.048248   \n",
       "pow_max                                  0.393946            0.309647   \n",
       "churn                                    0.012949           -0.002558   \n",
       "\n",
       "                                forecast_discount_energy  \\\n",
       "cons_12m                                       -0.043282   \n",
       "cons_gas_12m                                   -0.012595   \n",
       "cons_last_month                                -0.040874   \n",
       "forecast_cons_12m                               0.058435   \n",
       "forecast_cons_year                              0.008518   \n",
       "forecast_discount_energy                        1.000000   \n",
       "forecast_meter_rent_12m                        -0.008388   \n",
       "forecast_price_energy_off_peak                  0.353735   \n",
       "forecast_price_energy_peak                      0.059318   \n",
       "forecast_price_pow_off_peak                     0.048024   \n",
       "imp_cons                                        0.037939   \n",
       "margin_gross_pow_ele                            0.241142   \n",
       "margin_net_pow_ele                              0.240421   \n",
       "nb_prod_act                                     0.106812   \n",
       "net_margin                                      0.079784   \n",
       "num_years_antig                                -0.070697   \n",
       "pow_max                                        -0.005032   \n",
       "churn                                           0.017026   \n",
       "\n",
       "                                forecast_meter_rent_12m  \\\n",
       "cons_12m                                       0.065268   \n",
       "cons_gas_12m                                   0.041393   \n",
       "cons_last_month                                0.057476   \n",
       "forecast_cons_12m                              0.305627   \n",
       "forecast_cons_year                             0.276009   \n",
       "forecast_discount_energy                      -0.008388   \n",
       "forecast_meter_rent_12m                        1.000000   \n",
       "forecast_price_energy_off_peak                -0.579353   \n",
       "forecast_price_energy_peak                     0.706376   \n",
       "forecast_price_pow_off_peak                   -0.203089   \n",
       "imp_cons                                       0.224212   \n",
       "margin_gross_pow_ele                           0.183509   \n",
       "margin_net_pow_ele                             0.183566   \n",
       "nb_prod_act                                    0.018109   \n",
       "net_margin                                     0.275830   \n",
       "num_years_antig                                0.071558   \n",
       "pow_max                                        0.618549   \n",
       "churn                                          0.044245   \n",
       "\n",
       "                                forecast_price_energy_off_peak  \\\n",
       "cons_12m                                             -0.007748   \n",
       "cons_gas_12m                                         -0.017684   \n",
       "cons_last_month                                      -0.005187   \n",
       "forecast_cons_12m                                    -0.135646   \n",
       "forecast_cons_year                                   -0.158012   \n",
       "forecast_discount_energy                              0.353735   \n",
       "forecast_meter_rent_12m                              -0.579353   \n",
       "forecast_price_energy_off_peak                        1.000000   \n",
       "forecast_price_energy_peak                           -0.330138   \n",
       "forecast_price_pow_off_peak                           0.630377   \n",
       "imp_cons                                             -0.092535   \n",
       "margin_gross_pow_ele                                  0.094498   \n",
       "margin_net_pow_ele                                    0.094216   \n",
       "nb_prod_act                                           0.035093   \n",
       "net_margin                                           -0.135107   \n",
       "num_years_antig                                      -0.168095   \n",
       "pow_max                                              -0.384397   \n",
       "churn                                                -0.010837   \n",
       "\n",
       "                                forecast_price_energy_peak  \\\n",
       "cons_12m                                          0.145908   \n",
       "cons_gas_12m                                      0.074002   \n",
       "cons_last_month                                   0.136802   \n",
       "forecast_cons_12m                                 0.254056   \n",
       "forecast_cons_year                                0.251005   \n",
       "forecast_discount_energy                          0.059318   \n",
       "forecast_meter_rent_12m                           0.706376   \n",
       "forecast_price_energy_off_peak                   -0.330138   \n",
       "forecast_price_energy_peak                        1.000000   \n",
       "forecast_price_pow_off_peak                      -0.242017   \n",
       "imp_cons                                          0.212317   \n",
       "margin_gross_pow_ele                              0.165483   \n",
       "margin_net_pow_ele                                0.165560   \n",
       "nb_prod_act                                       0.044141   \n",
       "net_margin                                        0.234942   \n",
       "num_years_antig                                   0.080260   \n",
       "pow_max                                           0.411532   \n",
       "churn                                             0.029315   \n",
       "\n",
       "                                forecast_price_pow_off_peak  imp_cons  \\\n",
       "cons_12m                                          -0.026566  0.159711   \n",
       "cons_gas_12m                                      -0.020558  0.077846   \n",
       "cons_last_month                                   -0.023017  0.187034   \n",
       "forecast_cons_12m                                 -0.018477  0.634600   \n",
       "forecast_cons_year                                -0.044190  0.969395   \n",
       "forecast_discount_energy                           0.048024  0.037939   \n",
       "forecast_meter_rent_12m                           -0.203089  0.224212   \n",
       "forecast_price_energy_off_peak                     0.630377 -0.092535   \n",
       "forecast_price_energy_peak                        -0.242017  0.212317   \n",
       "forecast_price_pow_off_peak                        1.000000 -0.011318   \n",
       "imp_cons                                          -0.011318  1.000000   \n",
       "margin_gross_pow_ele                              -0.061869 -0.002099   \n",
       "margin_net_pow_ele                                -0.061925 -0.002084   \n",
       "nb_prod_act                                       -0.005335  0.046166   \n",
       "net_margin                                        -0.072623  0.458502   \n",
       "num_years_antig                                   -0.120136  0.024947   \n",
       "pow_max                                           -0.095897  0.260808   \n",
       "churn                                              0.014778 -0.001583   \n",
       "\n",
       "                                margin_gross_pow_ele  margin_net_pow_ele  \\\n",
       "cons_12m                                   -0.011926           -0.011905   \n",
       "cons_gas_12m                                0.006868            0.006852   \n",
       "cons_last_month                            -0.011498           -0.011477   \n",
       "forecast_cons_12m                          -0.023220           -0.023231   \n",
       "forecast_cons_year                         -0.010759           -0.010727   \n",
       "forecast_discount_energy                    0.241142            0.240421   \n",
       "forecast_meter_rent_12m                     0.183509            0.183566   \n",
       "forecast_price_energy_off_peak              0.094498            0.094216   \n",
       "forecast_price_energy_peak                  0.165483            0.165560   \n",
       "forecast_price_pow_off_peak                -0.061869           -0.061925   \n",
       "imp_cons                                   -0.002099           -0.002084   \n",
       "margin_gross_pow_ele                        1.000000            0.999914   \n",
       "margin_net_pow_ele                          0.999914            1.000000   \n",
       "nb_prod_act                                -0.006784           -0.006763   \n",
       "net_margin                                  0.031814            0.031639   \n",
       "num_years_antig                            -0.061091           -0.061197   \n",
       "pow_max                                     0.373298            0.373351   \n",
       "churn                                       0.095725            0.095772   \n",
       "\n",
       "                                nb_prod_act  net_margin  num_years_antig  \\\n",
       "cons_12m                           0.154251    0.133614        -0.003565   \n",
       "cons_gas_12m                       0.239387    0.070867        -0.013815   \n",
       "cons_last_month                    0.169099    0.121835        -0.003677   \n",
       "forecast_cons_12m                  0.055915    0.768537         0.021283   \n",
       "forecast_cons_year                 0.040426    0.464928         0.048248   \n",
       "forecast_discount_energy           0.106812    0.079784        -0.070697   \n",
       "forecast_meter_rent_12m            0.018109    0.275830         0.071558   \n",
       "forecast_price_energy_off_peak     0.035093   -0.135107        -0.168095   \n",
       "forecast_price_energy_peak         0.044141    0.234942         0.080260   \n",
       "forecast_price_pow_off_peak       -0.005335   -0.072623        -0.120136   \n",
       "imp_cons                           0.046166    0.458502         0.024947   \n",
       "margin_gross_pow_ele              -0.006784    0.031814        -0.061091   \n",
       "margin_net_pow_ele                -0.006763    0.031639        -0.061197   \n",
       "nb_prod_act                        1.000000    0.053781         0.005228   \n",
       "net_margin                         0.053781    1.000000        -0.000263   \n",
       "num_years_antig                    0.005228   -0.000263         1.000000   \n",
       "pow_max                            0.012712    0.325417         0.059715   \n",
       "churn                             -0.014930    0.041135        -0.074140   \n",
       "\n",
       "                                 pow_max     churn  \n",
       "cons_12m                        0.082889 -0.045968  \n",
       "cons_gas_12m                    0.054317 -0.037957  \n",
       "cons_last_month                 0.074529 -0.045284  \n",
       "forecast_cons_12m               0.393946  0.012949  \n",
       "forecast_cons_year              0.309647 -0.002558  \n",
       "forecast_discount_energy       -0.005032  0.017026  \n",
       "forecast_meter_rent_12m         0.618549  0.044245  \n",
       "forecast_price_energy_off_peak -0.384397 -0.010837  \n",
       "forecast_price_energy_peak      0.411532  0.029315  \n",
       "forecast_price_pow_off_peak    -0.095897  0.014778  \n",
       "imp_cons                        0.260808 -0.001583  \n",
       "margin_gross_pow_ele            0.373298  0.095725  \n",
       "margin_net_pow_ele              0.373351  0.095772  \n",
       "nb_prod_act                     0.012712 -0.014930  \n",
       "net_margin                      0.325417  0.041135  \n",
       "num_years_antig                 0.059715 -0.074140  \n",
       "pow_max                         1.000000  0.030362  \n",
       "churn                           0.030362  1.000000  "
      ]
     },
     "execution_count": 94,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfc_num.corr().T"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "6f92a544",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Numerical columns having maximum corelation with churn:\n",
    "#•\tMargin_gross_pow_ele\n",
    "#•\tmargin_net_pow_ele\n",
    "#•\tnum_years_antig\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "d24db94d",
   "metadata": {},
   "outputs": [],
   "source": [
    "dfp=price_df.copy()  #Store a copy of it"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "4e2484c1",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "id                    0\n",
       "price_date            0\n",
       "price_off_peak_var    0\n",
       "price_peak_var        0\n",
       "price_mid_peak_var    0\n",
       "price_off_peak_fix    0\n",
       "price_peak_fix        0\n",
       "price_mid_peak_fix    0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dfp.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "1df58e3a",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Datatype convertion\n",
    "dfp.price_date=dfp.price_date.astype(\"datetime64[ns]\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "eba67489",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((14606, 26), (193002, 8))"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    " dfc.shape,dfp.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "9b8e2654",
   "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>id</th>\n",
       "      <th>price_date</th>\n",
       "      <th>price_off_peak_var</th>\n",
       "      <th>price_peak_var</th>\n",
       "      <th>price_mid_peak_var</th>\n",
       "      <th>price_off_peak_fix</th>\n",
       "      <th>price_peak_fix</th>\n",
       "      <th>price_mid_peak_fix</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>2015-01-01</td>\n",
       "      <td>0.151367</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>44.266931</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>2015-02-01</td>\n",
       "      <td>0.151367</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>44.266931</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>2015-03-01</td>\n",
       "      <td>0.151367</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>44.266931</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>2015-04-01</td>\n",
       "      <td>0.149626</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>44.266931</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>2015-05-01</td>\n",
       "      <td>0.149626</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>44.266931</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 id price_date  price_off_peak_var  \\\n",
       "0  038af19179925da21a25619c5a24b745 2015-01-01            0.151367   \n",
       "1  038af19179925da21a25619c5a24b745 2015-02-01            0.151367   \n",
       "2  038af19179925da21a25619c5a24b745 2015-03-01            0.151367   \n",
       "3  038af19179925da21a25619c5a24b745 2015-04-01            0.149626   \n",
       "4  038af19179925da21a25619c5a24b745 2015-05-01            0.149626   \n",
       "\n",
       "   price_peak_var  price_mid_peak_var  price_off_peak_fix  price_peak_fix  \\\n",
       "0             0.0                 0.0           44.266931             0.0   \n",
       "1             0.0                 0.0           44.266931             0.0   \n",
       "2             0.0                 0.0           44.266931             0.0   \n",
       "3             0.0                 0.0           44.266931             0.0   \n",
       "4             0.0                 0.0           44.266931             0.0   \n",
       "\n",
       "   price_mid_peak_fix  \n",
       "0                 0.0  \n",
       "1                 0.0  \n",
       "2                 0.0  \n",
       "3                 0.0  \n",
       "4                 0.0  "
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Creating grouped feature for each company id.\n",
    "dfp.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "65068e39",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA3cAAAHiCAYAAABLBzXPAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuNCwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8QVMy6AAAACXBIWXMAAAsTAAALEwEAmpwYAAAl0UlEQVR4nO3de5DddX3/8dfeAoEgl7gRLJRCzKCDBXVGBLlktMNK2Sw3EYMRKlaNBRK1WkrWtLS1XEztZIjCONaYSuhPEwOBgIA6ocJgQBSUGIoEkA2h4RdilIXUkN3snt8f/LrTVFhO2D17dj/7eMxkJuecPef7DjPvbJ58v+dsQ6VSqQQAAIAxrbHeAwAAADB04g4AAKAA4g4AAKAA4g4AAKAA4g4AAKAA4g4AAKAAzfUeYHf99rf/lf5+P70BAAAYXxobG7L//nu/4uNjLu76+yviDgAA4H9xWSYAAEABxB0AAEABxB0AAEABxB0AAEABxB0AAEABxB0AAEABxB0AAEABxB0AAEABxB0AAEABxB0AAEABxB0AAEABxB0AAEABxB0AAEABxB0AAEABxB0AAEABxB0AAEABmus9AK/uRz+6O/fcc1e9x3hV3d3PJUn23Xe/us7xak44YXqOP/6keo/BCLJDw88ejS92aPjZofHFDg0/O/TynLlj2HR3d6e7u7veY8CYZYdgaOwQDI0dGvsaKpVKpd5D7I6tW7elv39MjTxufPGLX0iS/PVf/02dJ4GxyQ7B0NghGBo7NPo1NjZk8uRJr/z4CM4CAABAjYg7AACAAog7AACAAog7AACAAog7AACAAog7AACAAog7AACAAog7AACAAog7AACAAog7AACAAog7AACAAjTX8sWvvvrqfO9730tDQ0POPvvsXHDBBZk3b14eeOCBTJw4MUly8cUX5+STT67lGAAAAMWrWdzdf//9ue+++7Jq1ars3Lkzp556aqZPn55169bl+uuvz5QpU2p1aAAAgHGnZpdlHnPMMbnuuuvS3NycrVu3pq+vL3vuuWc2bdqUzs7OdHR0ZNGiRenv76/VCAAAAONGTS/LbGlpyaJFi/KNb3wjp5xySnbu3Jljjz02l112WfbZZ5/Mnj07K1asyDnnnFP1a06ePKmGEzMULS1NSZLW1n3qPAmMTXYIhsYOwdDYobGvpnGXJHPnzs3HP/7xfPKTn8y9996ba665ZuCx8847LzfddNNuxd3WrdvS31+pxagMUW9vX5Jky5YX6jwJjE12CIbGDsHQ2KHRr7GxYdCTXTW7LPOJJ57II488kiSZOHFi2tractttt+V73/vewNdUKpU0N9e8LwEAAIpXs7h7+umnM3/+/PT09KSnpyerV6/OO9/5zlxxxRXp7u5Ob29vli1b5pMyAQAAhkHNTptNnz49a9euzRlnnJGmpqa0tbXl4osvzv77759zzz03O3fuTFtbW2bMmFGrEQAAAMaNml4TOWfOnMyZM2eX+2bNmpVZs2bV8rAAAADjTs0uywQAAGDkiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIAC1DTurr766px66qlpb2/PkiVLkiRr1qxJR0dH2trasnDhwloeHgAAYNxortUL33///bnvvvuyatWq7Ny5M6eeemqOO+64dHZ2ZunSpTnooIMye/bs3HXXXZk+fXqtxgAAABgXanbm7phjjsl1112X5ubmbN26NX19fXn++edz6KGH5pBDDklzc3M6Ojpyxx131GoEAACAcaOml2W2tLRk0aJFaW9vz3HHHZdnn302ra2tA49PmTIlmzdvruUIAAAA40LNLsv8b3Pnzs3HP/7xfPKTn0xXV1caGhoGHqtUKrvcrsbkyZOGe0SGSUtLU5KktXWfOk8CY5MdgqGxQzA0dmjsq1ncPfHEE+np6clb3vKWTJw4MW1tbbnjjjvS1NQ08DVbtmzJlClTdut1t27dlv7+ynCPyzDo7e1LkmzZ8kKdJ4GxyQ7B0NghGBo7NPo1NjYMerKrZpdlPv3005k/f356enrS09OT1atXZ+bMmXnyySezYcOG9PX15dZbb81JJ51UqxEAAADGjZqduZs+fXrWrl2bM844I01NTWlra0t7e3sOOOCAzJkzJzt27Mj06dNzyimn1GoEAACAcaOm77mbM2dO5syZs8t9xx13XFatWlXLwwIAAIw7Nf20TAAAAEaGuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAACiAuAMAAChAcy1f/Ctf+Upuv/32JMn06dNzySWXZN68eXnggQcyceLEJMnFF1+ck08+uZZjAAAAFK9mcbdmzZrcc889WblyZRoaGvKxj30sP/jBD7Ju3bpcf/31mTJlSq0ODQAAMO7U7LLM1tbWXHrppZkwYUJaWloyderUbNq0KZs2bUpnZ2c6OjqyaNGi9Pf312oEAACAcaNmcTdt2rS87W1vS5J0dXXl9ttvz4knnphjjz02V1xxRZYvX56f/vSnWbFiRa1GAAAAGDdq+p67JHnssccye/bsXHLJJTn88MNzzTXXDDx23nnn5aabbso555xT9etNnjypFmMyDFpampIkra371HkSGJvsEAyNHYKhsUNjX03j7oEHHsjcuXPT2dmZ9vb2PProo+nq6sr73ve+JEmlUklz8+6NsHXrtvT3V2oxLkPU29uXJNmy5YU6TwJjkx2CobFDMDR2aPRrbGwY9GRXzS7LfOaZZ3LRRRflS1/6Utrb25O8FHNXXHFFuru709vbm2XLlvmkTAAAgGFQszN3ixcvzo4dO3LVVVcN3Ddz5sx84hOfyLnnnpudO3emra0tM2bMqNUIAAAA40bN4m7+/PmZP3/+yz42a9asWh0WAABgXKr5B6oAAC/v//yf67Jx44Z6j1GMp5566b/lF7/4hTpPUo5DDjk0H/rQ+fUeY1D2aPjYoeE30js0ruPOXwbDy18Iw8s31PHHDg2/0b5HGzduyPpfPZqmfSfUe5Qi9De99GEQT2x9ss6TlKGvu6feI1Rl48YNeXL9L/P6pqZ6jzLm7fH/f/70C088VudJyvDrvr4RP+a4jruNGzfk0cceT9Oe+9V7lCL09730l+rjG39d50nGvr4Xn6v3CFXZuHFDuh7/ZQ6cNK7/Khk2e+Wlb6ov/t/H6zxJGf7vtp31HqEqTftOyL4nvbHeY8Dv6b57U71HqNrrm5py+j771XsM2MXNLzw34scc9/8ia9pzv+x16J/UewzYxe82rK73CFU7cFJzLjjqgHqPAb9nydrf1HsEABhRNftRCAAAAIwccQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFAAcQcAAFCA1xx3/f39wzkHAAAAQ1BV3M2fPz89PT0Dt5955pnMmjWrZkMBAACwe6qKu56enpxzzjnZuHFj7rjjjnzgAx/In/zJn9R6NgAAAKrUXM0XLViwICtWrMhpp52WSZMmZcmSJZk2bVqtZwMAAKBKVZ25++Uvf5mlS5fmxBNPzOTJk/Mv//Iv2bZtW61nAwAAoEpVxd3555+fD3/4w1m0aFGWL1+e173udeno6Kj1bAAAAFSpqssyv/Wtb2Xq1KlJkgkTJmT+/Pk58cQTazoYAAAA1Rs07m6++eacfvrpufvuu3P33Xf/3uPTp0+v2WAAAABUb9C4e+qpp5Ik69evH5FhAAAAeG0GjbuHHnooSfLmN785f/ZnfzYiAwEAALD7Bo27J554IrfcckuWLl2aN77xjalUKrs83tbWVtPhAAAAqM6gcTd37tysWLEiW7duzXXXXbfLYw0NDeIOAABglBg07s4888yceeaZufLKKzNv3ryX/Zpbb701M2bMqMlwAAAAVKeqn3P3SmGXJIsXLx62YQAAAHhtqoq7wfzv9+EBAAAw8oYcdw0NDa/42Fe+8pW0t7envb09CxYsSJKsWbMmHR0daWtry8KFC4d6eAAAADIMcfdK1qxZk3vuuScrV67MTTfdlIcffji33nprOjs7c+211+a2227LunXrctddd9VqBAAAgHGjZnHX2tqaSy+9NBMmTEhLS0umTp2arq6uHHrooTnkkEPS3Nycjo6O3HHHHbUaAQAAYNwY9NMyq/FK77mbNm3awO+7urpy++2358Mf/nBaW1sH7p8yZUo2b968W8ebPHnSaxv0ZbS0NA3ba8Fwa2lpSmvrPvUeY1AtLU15sd5DwCBG+x75PsRoN9p3KLFHjG4jvUNVxd327dszceLEXe77+c9/nre97W3p6OgY9LmPPfZYZs+enUsuuSRNTU3p6uoaeKxSqQz6nr2Xs3XrtvT3D8+HuPT29g3L60At9Pb2ZcuWF+o9xqDsEKPdaN8jO8RoN9p3KLFHjG7DvUONjQ2Dnuyq6rLMCy+8ML29vUmSvr6+LFy4MLNnz06S/Pmf//krPu+BBx7IRz7ykXz2s5/NmWeemQMPPDBbtmwZeHzLli2ZMmVKVX8QAAAAXllVcTdt2rR89rOfzWOPPZYPfOADWbt2bVauXDnoc5555plcdNFF+dKXvpT29vYkydFHH50nn3wyGzZsSF9fX2699dacdNJJQ/9TAAAAjHNVXZbZ2dmZL3zhCznjjDPS2dmZWbNmvepzFi9enB07duSqq64auG/mzJm56qqrMmfOnOzYsSPTp0/PKaec8tqnBwAAIMmrxN2SJUsGfv/GN74x+++/fx588MH09PQkSS644IJXfO78+fMzf/78l31s1apVr2VWAAAAXsGgcbd+/fpdbp944okvez8AAAD1NWjcXXnllSM1BwAAAENQ1Xvufvazn+VrX/tafve736VSqaS/vz9PP/10fvjDH9Z4PAAAAKpR1adlzp8/P29/+9uzbdu2dHR0ZNKkSWlra6v1bAAAAFSpqjN3DQ0N+cQnPpHf/va3Ofzww9PR0ZH3v//9tZ4NAACAKlV15m7vvfdOkvzhH/5hHnvssey5555pbKzqqQAAAIyAqs7cHXXUUfn0pz+dT33qU5k9e3a6urrS3FzVUwEAABgBVZ1+6+zszEc+8pEcdthh6ezsTH9/f/75n/+51rMBAABQparfc9fY2Jhvf/vbOeuss7Lvvvvm8MMPr/VsAAAAVKmqM3c33HBD5s2bl69//et54YUXcuGFF2b58uW1ng0AAIAqVRV3119/fZYtW5ZJkyZl8uTJufHGG/PNb36z1rMBAABQparirrGxMZMmTRq4fdBBB6WpqalmQwEAALB7qoq7/fbbL4888kgaGhqSJKtWrcq+++5b08EAAACoXlUfqNLZ2ZlPfepTeeqpp3LCCSdkjz32yLXXXlvr2QAAAKhSVXE3derU3Hzzzenq6kpfX18OO+ywtLS01Ho2AAAAqlRV3PX19eXb3/527rnnnjQ1NeW9731vzjrrrFrPBgAAQJWqirt//Md/zOOPP57TTz89lUolK1asyIYNG/KZz3ym1vMBAABQhari7kc/+lG++93vDlyKedppp+W0004TdwAAAKNEVZ+WecABB6Svr2/gdkNDQ173utfVbCgAAAB2T1Vn7t785jfnQx/6UM4666w0NTXltttuy/77758lS5YkSS644IKaDgkAAMDgqoq7HTt25IgjjsjDDz+cJDn44IOTJOvXr6/dZAAAAFStqri78sorX/Gxv/zLvxy2YQAAAHhtqnrP3WCefPLJ4ZgDAACAIRhy3AEAAFB/4g4AAKAA4g4AAKAA4g4AAKAAQ467SqUyHHMAAAAwBFXH3dq1a/Ptb387PT09+dnPfjZw/8KFC2syGAAAANWrKu5uvPHGzJs3L1//+tfzwgsv5MILL8zy5cuTJIcddlhNBwQAAODVVRV3S5cuzbJlyzJp0qRMnjw5N954Y775zW/WejYAAACqVFXcNTY2ZtKkSQO3DzrooDQ1NdVsKAAAAHZPVXG333775ZFHHklDQ0OSZNWqVdl3331rOhgAAADVa67mizo7O/OpT30qTz31VE444YTsscceufbaa2s9GwAAAFWqKu6mTp2alStX5qmnnkpfX18OP/zwNDdX9VQAAABGQFWXZd533315//vfn6lTp6ahoSHTp0/f5cchAAAAUF9Vxd2CBQty5ZVXJkmmTZuWr33tawO3AQAAqL+q4q63tzdHHnnkwO0jjzwyPT09NRsKAACA3VNV3E2cODF33333wO177703e+21V82GAgAAYPdU9akon//853PRRRcNfIhKY2NjvvzlL9d0MAAAAKpXVdwdffTR+eEPf5j169enqakphx12WCZMmFDr2QAAAKjSoHF388035/TTT8+SJUt2uX/NmjVJkgsuuKB2kwEAAFC1QeNuw4YNSZL169ePyDAAAAC8NoPG3dy5c5Mkr3/96/PZz352RAYCAABg91X1aZk//OEPazwGAAAAQ1HVB6ocfPDB+ehHP5p3vOMd2XvvvQfuf7X33G3bti0zZ87MV7/61Rx88MGZN29eHnjggUycODFJcvHFF+fkk08ewvgAAAAkVcbdfvvtlyR5+OGH09TUlH322edVn/PQQw9l/vz56erqGrhv3bp1uf766zNlypTXNCwAAAAvr6rLMj/2sY9l/fr1ueuuu3LnnXdm48aNmTNnzqDPWb58eS677LKBkNu+fXs2bdqUzs7OdHR0ZNGiRenv7x/6nwAAAIDqztx1dnbmnHPOyVlnnZVKpZJly5bl85///O/9iIT/6fLLL9/l9q9//esce+yxueyyy7LPPvtk9uzZWbFiRc4555zdGnjy5Em79fWDaWlpGrbXguHW0tKU1tZXP0teTy0tTXmx3kPAIEb7Hvk+xGg32ncosUeMbiO9Q1XF3fbt2/PBD35w4PZ5552X5cuX79aBDjnkkFxzzTW7vMZNN92023G3deu29PdXdus5r6S3t29YXgdqobe3L1u2vFDvMQZlhxjtRvse2SFGu9G+Q4k9YnQb7h1qbGwY9GRXVZdlHn744XnwwQcHbq9fvz4HH3zwbg3y6KOP5nvf+97A7UqlkubmqtoSAACAV1FVXW3atCnnnXdejjjiiDQ3N+c//uM/0tramo6OjiTJLbfc8qqvUalUcsUVV+TYY4/NXnvtlWXLluXMM88c2vQAAAAkqTLuPve5zw35QG9+85vziU98Iueee2527tyZtra2zJgxY8ivCwAAQJVxd8wxx7zmA9x5550Dv581a1ZmzZr1ml8LAACAl1fVe+4AAAAY3cQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAcQdAABAAWoad9u2bcuMGTPy9NNPJ0nWrFmTjo6OtLW1ZeHChbU8NAAAwLhSs7h76KGHcu6556arqytJ8uKLL6azszPXXnttbrvttqxbty533XVXrQ4PAAAwrtQs7pYvX57LLrssU6ZMSZKsXbs2hx56aA455JA0Nzeno6Mjd9xxR60ODwAAMK401+qFL7/88l1uP/vss2ltbR24PWXKlGzevHm3X3fy5ElDnu2/tbQ0DdtrwXBraWlKa+s+9R5jUC0tTXmx3kPAIEb7Hvk+xGg32ncosUeMbiO9QzWLu/+tv78/DQ0NA7crlcout6u1deu29PdXhmWm3t6+YXkdqIXe3r5s2fJCvccYlB1itBvte2SHGO1G+w4l9ojRbbh3qLGxYdCTXSP2aZkHHnhgtmzZMnB7y5YtA5dsAgAAMDQjFndHH310nnzyyWzYsCF9fX259dZbc9JJJ43U4QEAAIo2Ypdl7rHHHrnqqqsyZ86c7NixI9OnT88pp5wyUocHAAAoWs3j7s477xz4/XHHHZdVq1bV+pAAAADjzohdlgkAAEDtiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACiDsAAIACNNfjoOedd15+85vfpLn5pcP/wz/8Q44++uh6jAIAAFCEEY+7SqWSrq6u/Pu///tA3AEAADA0I35Z5q9+9askyUc/+tGcdtppuf7660d6BAAAgOKM+Kmz559/Pscdd1z+5m/+Jr29vTn//PNz2GGH5fjjj6/q+ZMnTxq2WVpamobttWC4tbQ0pbV1n3qPMaiWlqa8WO8hYBCjfY98H2K0G+07lNgjRreR3qERj7u3v/3tefvb3z5w++yzz85dd91Vddxt3bot/f2VYZmlt7dvWF4HaqG3ty9btrxQ7zEGZYcY7Ub7HtkhRrvRvkOJPWJ0G+4damxsGPRk14hflvnTn/40995778DtSqXivXcAAABDNOJx98ILL2TBggXZsWNHtm3blpUrV+bkk08e6TEAAACKMuKnzN7znvfkoYceyhlnnJH+/v586EMf2uUyTQAAAHZfXa6H/PSnP51Pf/rT9Tg0AABAkUb8skwAAACGn7gDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAogLgDAAAoQF3i7pZbbsmpp56atra2/Nu//Vs9RgAAAChK80gfcPPmzVm4cGFuvPHGTJgwITNnzsy73vWuvOlNbxrpUQAAAIox4mfu1qxZk2OPPTb77bdf9tprr7zvfe/LHXfcMdJjAAAAFGXEz9w9++yzaW1tHbg9ZcqUrF27dqTHSJJ0dz+Xvt/9Oi88ekNdjl+1Sv9LvxgeDY0v/RrN+nemu3vE13O3dXc/l83dvblyzbP1HmVQfZVK+qzQsGpqTJoaGuo9xqB6+ip5w8Tn6j3GoLq7n8vOrS9m66on6z3K4PqT9FfqPUVZGhtG/ycf9FXS3fxcvad4Vd3dz2Xzzp1Z/Ntf13uUV9SXl9aI4dOYpKneQ7yK3iT93c+N6DFH/F+P/f39afgf/yCoVCq73H41kydPGrZZDjrowDz/fPewvV6t7Ny5Mzt3+qY6XJqbm9LcPNrDqSUHHXRgWlv3qfcggxorO1TZuTP9O3fWe4yiNDY3p2mU79HEZNTv0VjZoZe+D9mh4dTc3DwGvheN/h1Kxsge7dyZih0aVk1jYIeaM/I71FCpVEa0GlauXJmf/vSnufzyy5Mk11xzTSqVSi6++OKqnr9167b0+7+HAADAONPY2DDoya4RvyDg3e9+d+6999785je/yfbt2/P9738/J5100kiPAQAAUJQRP5f5hje8IZ/5zGdy/vnnp7e3N2effXaOOuqokR4DAACgKCN+WeZQuSwTAAAYj0bdZZkAAAAMP3EHAABQAHEHAABQAHEHAABQAHEHAABQAHEHAABQAHEHAABQAHEHAABQAHEHAABQAHEHAABQAHEHAABQAHEHAABQAHEHAABQgOZ6D7C7Ghsb6j0CAADAiHu1FmqoVCqVEZoFAACAGnFZJgAAQAHEHQAAQAHEHQAAQAHEHQAAQAHEHQAAQAHEHQAAQAHEHQAAQAHEHQAAQAHEHQAAQAHEHQAAQAHEHcNm27ZtmTFjRp5++ul6jwJjzle+8pW0t7envb09CxYsqPc4MOZcffXVOfXUU9Pe3p4lS5bUexwYs774xS/m0ksvrfcYvEbijmHx0EMP5dxzz01XV1e9R4ExZ82aNbnnnnuycuXK3HTTTXn44Yfzgx/8oN5jwZhx//3357777suqVatyww03ZOnSpfnVr35V77FgzLn33nuzcuXKeo/BEIg7hsXy5ctz2WWXZcqUKfUeBcac1tbWXHrppZkwYUJaWloyderUbNq0qd5jwZhxzDHH5Lrrrktzc3O2bt2avr6+7LXXXvUeC8aU5557LgsXLswnP/nJeo/CEDTXewDKcPnll9d7BBizpk2bNvD7rq6u3H777fnWt75Vx4lg7GlpacmiRYvyjW98I6ecckre8IY31HskGFP+9m//Np/5zGfyzDPP1HsUhsCZO4BR4rHHHstHP/rRXHLJJfmjP/qjeo8DY87cuXNz77335plnnsny5cvrPQ6MGd/5zndy0EEH5bjjjqv3KAyRM3cAo8ADDzyQuXPnprOzM+3t7fUeB8aUJ554Ij09PXnLW96SiRMnpq2tLY8++mi9x4Ix47bbbsuWLVty+umnp7u7O7/73e9yxRVXpLOzs96jsZvEHUCdPfPMM7nooouycOFC/9cUXoOnn346ixYtGricefXq1Xn/+99f56lg7PifnzB744035v777xd2Y5S4A6izxYsXZ8eOHbnqqqsG7ps5c2bOPffcOk4FY8f06dOzdu3anHHGGWlqakpbW5sz4MC41FCpVCr1HgIAAICh8YEqAAAABRB3AAAABRB3AAAABRB3AAAABRB3AAAABRB3ABRv586d+drXvpaOjo50dHSkvb09f//3f5/f/va3w3aMjRs3Zs6cOcP2egCwu/ycOwCK91d/9Vfp7+/P9ddfn3333Te9vb3513/918ycOTM33HBDJk2aNORjbNq0KU8++eQwTAsAr42fcwdA0dauXZsLL7wwP/jBDzJx4sRdHvuLv/iLvOMd78i3vvWtXH311fnjP/7jJMl73/vegdtf/epXs3r16rz44ovZvn17/vqv/zonn3xyvvzlL+fnP/95nn322UybNi2/+MUvsnnz5rzzne/M4sWL8+CDD+ZLX/pStm/fnsbGxlx88cV5z3vekxtvvDErVqzI9u3bM2nSpCxdurQe/1kAKJAzdwAU7cEHH8xb3/rW3wu7JDn++OPzox/96BWf+5//+Z9Zs2ZNli5dmj333DPf/e53s2jRopx88skDj996661pbm7Oj3/843zhC1/I4sWL093dnXnz5mXx4sU5+OCDs3nz5pxzzjk54ogjkiSPP/547rzzzmE5YwgA/03cATCuDXYByx/8wR9kwYIFueWWW7Jhw4Y89NBD+a//+q+Bx9/2trelufn3v5X+/Oc/z5YtW3LRRRcN3NfQ0JBHH300SXLEEUcIOwCGnQ9UAaBo73jHO/KLX/wi27dvT5L09PQMfJDKfffdl6OPPjrJrpHX09OTJHn44YfzwQ9+MNu2bcvxxx+fj33sY7u89l577fWyx+zr68vUqVNz8803D/xatmxZTjjhhEGfBwBDIe4AKNpRRx2Vd73rXbn00kvT3d2djRs3ZtasWZkzZ04effTRzJo1KwcccEDWrVuXJPnxj3+cLVu2JEl+8pOf5K1vfWsuuOCCHHPMMVm9enX6+vpe9jhNTU3p7e1N8tIZvQ0bNuQnP/lJkuSRRx7J+973vmzevHkE/sQAjFcuywSgeP/0T/+Ub3zjG/nwhz+cJOnt7U1TU1P23nvvrF69Op/73Ofyd3/3d1m2bFmOPPLIHHnkkUmSGTNm5Pvf/37+9E//NP39/XnPe96T7u7ubNu27feO8aY3vSl77LFHzj777HznO9/JokWLsmDBguzYsSOVSiULFizIwQcfnPvvv39E/+wAjB8+LROAcev555/PunXr8u53v7veowDAkIk7AACAAnjPHQAAQAHEHQAAQAHEHQAAQAHEHQAAQAHEHQAAQAHEHQAAQAH+H4KsKGYr+QBqAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "#Lets observe quarter wise peak, off-peak, and mid-peak consumption users quarter wise\n",
    "\n",
    "dfp['Quarter']=dfp.price_date.apply(lambda x:x.quarter)\n",
    "qt_df=dfp.groupby(['id','Quarter'],as_index=False).mean()\n",
    "\n",
    "for i in qt_df.columns[2:]:\n",
    "    plt.figure()\n",
    "    sns.boxplot(x=qt_df.Quarter,y=qt_df[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "ed1a7969",
   "metadata": {},
   "outputs": [],
   "source": [
    "### It seems like there is no variance in pricing with respect to quarter of year.\n",
    "### As the data is for year 2015 only, so there wont be any year-wise comparision."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "id": "953063cc",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Lets drop quarter column as it has become insignificant for our analysis\n",
    "\n",
    "dfp.drop('Quarter',1,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "id": "39778438",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Create a month column for price_date.\n",
    "dfp['Month']=dfp.price_date.apply(lambda x:x.month)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "a8565176",
   "metadata": {},
   "outputs": [],
   "source": [
    "#It seems like the data is sorted according to customer and moths"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "ecb76a95",
   "metadata": {},
   "source": [
    "### Lets define price sensitivity as difference between next month and present month divided by present month price."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "25ace607",
   "metadata": {
    "scrolled": false
   },
   "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>id</th>\n",
       "      <th>price_off_peak_var</th>\n",
       "      <th>price_peak_var</th>\n",
       "      <th>price_mid_peak_var</th>\n",
       "      <th>price_off_peak_fix</th>\n",
       "      <th>price_peak_fix</th>\n",
       "      <th>price_mid_peak_fix</th>\n",
       "      <th>Month</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>-1.150185</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.000003</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>0.464491</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.401609</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>-2.968314</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>31f2ce549924679a3cbb2d128ae9ea43</td>\n",
       "      <td>-13.631658</td>\n",
       "      <td>inf</td>\n",
       "      <td>inf</td>\n",
       "      <td>-8.727114</td>\n",
       "      <td>inf</td>\n",
       "      <td>inf</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>31f2ce549924679a3cbb2d128ae9ea43</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>31f2ce549924679a3cbb2d128ae9ea43</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  id  price_off_peak_var  price_peak_var  \\\n",
       "0   038af19179925da21a25619c5a24b745                 NaN             NaN   \n",
       "1   038af19179925da21a25619c5a24b745            0.000000             NaN   \n",
       "2   038af19179925da21a25619c5a24b745            0.000000             NaN   \n",
       "3   038af19179925da21a25619c5a24b745           -1.150185             NaN   \n",
       "4   038af19179925da21a25619c5a24b745            0.000000             NaN   \n",
       "5   038af19179925da21a25619c5a24b745            0.000000             NaN   \n",
       "6   038af19179925da21a25619c5a24b745            0.464491             NaN   \n",
       "7   038af19179925da21a25619c5a24b745           -2.968314             NaN   \n",
       "8   038af19179925da21a25619c5a24b745            0.000000             NaN   \n",
       "9   038af19179925da21a25619c5a24b745            0.000000             NaN   \n",
       "10  038af19179925da21a25619c5a24b745            0.000000             NaN   \n",
       "11  038af19179925da21a25619c5a24b745            0.000000             NaN   \n",
       "12  31f2ce549924679a3cbb2d128ae9ea43          -13.631658             inf   \n",
       "13  31f2ce549924679a3cbb2d128ae9ea43            0.000000             0.0   \n",
       "14  31f2ce549924679a3cbb2d128ae9ea43            0.000000             0.0   \n",
       "\n",
       "    price_mid_peak_var  price_off_peak_fix  price_peak_fix  \\\n",
       "0                  NaN                 NaN             NaN   \n",
       "1                  NaN            0.000000             NaN   \n",
       "2                  NaN            0.000000             NaN   \n",
       "3                  NaN            0.000000             NaN   \n",
       "4                  NaN            0.000000             NaN   \n",
       "5                  NaN           -0.000003             NaN   \n",
       "6                  NaN            0.401609             NaN   \n",
       "7                  NaN            0.000000             NaN   \n",
       "8                  NaN            0.000000             NaN   \n",
       "9                  NaN            0.000000             NaN   \n",
       "10                 NaN            0.000000             NaN   \n",
       "11                 NaN            0.000000             NaN   \n",
       "12                 inf           -8.727114             inf   \n",
       "13                 0.0            0.000000             0.0   \n",
       "14                 0.0            0.000000             0.0   \n",
       "\n",
       "    price_mid_peak_fix  Month  \n",
       "0                  NaN      1  \n",
       "1                  NaN      2  \n",
       "2                  NaN      3  \n",
       "3                  NaN      4  \n",
       "4                  NaN      5  \n",
       "5                  NaN      6  \n",
       "6                  NaN      7  \n",
       "7                  NaN      8  \n",
       "8                  NaN      9  \n",
       "9                  NaN     10  \n",
       "10                 NaN     11  \n",
       "11                 NaN     12  \n",
       "12                 inf      1  \n",
       "13                 0.0      2  \n",
       "14                 0.0      3  "
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "price_sen_per=(dfp.iloc[:,2:8].diff(1)/dfp.iloc[:,2:8].shift(1))*100\n",
    "price_sen_per=pd.concat([dfp['id'],price_sen_per,dfp['Month']],1)\n",
    "price_sen_per.head(15)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c22f538a",
   "metadata": {},
   "source": [
    "#Problem with this dataframe is for a different customer the difference is taken so when a new customer is encountred lets make\n",
    "#it null column as there is no previous data available for that."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "73a61fa9",
   "metadata": {},
   "source": [
    "el=list()\n",
    "for i in price_sen_per.index:\n",
    "    if i!=len(price_sen_per)-1:\n",
    "        if price_sen_per.id[i]!=price_sen_per.id[i+1]:\n",
    "            el.append(i+1)\n",
    "        \n"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "1efa9af0",
   "metadata": {},
   "source": [
    "earr=np.empty(6)\n",
    "earr[:]=np.nan\n",
    "for i in el:\n",
    "    price_sen_per.iloc[i,1:7]=earr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "5f65b944",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Storing it as a csv file\n",
    "#price_sen_per.to_csv('Price_Sensitivity.csv',index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "id": "71aa6fb4",
   "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>id</th>\n",
       "      <th>Month</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>86</th>\n",
       "      <td>0117022254b01c36da23c5329a97e653</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>171</th>\n",
       "      <td>02bc81ecebcc8328dba4bf07bd6b6d55</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>333</th>\n",
       "      <td>0580216fd8b3a562087d52673f6ba63d</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>493</th>\n",
       "      <td>07cf5cb6081706df0be8b5512bc18000</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>938</th>\n",
       "      <td>0e7e0352b2dc0c2d5576bd9daaa4c609</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                   id  Month\n",
       "86   0117022254b01c36da23c5329a97e653     11\n",
       "171  02bc81ecebcc8328dba4bf07bd6b6d55     11\n",
       "333  0580216fd8b3a562087d52673f6ba63d     11\n",
       "493  07cf5cb6081706df0be8b5512bc18000     10\n",
       "938  0e7e0352b2dc0c2d5576bd9daaa4c609     11"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Lets check for overall price effect on churn\n",
    "month_cnt=dfp.groupby(['id'],as_index=False)['Month'].count()\n",
    "month_cnt[month_cnt.Month!=12].head()\n",
    "# There are certain customers who doesnt have year long transaction.\n",
    "#Lets merge this data with a group by sum of data for further analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "f013a79f",
   "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>id</th>\n",
       "      <th>price_off_peak_var</th>\n",
       "      <th>price_peak_var</th>\n",
       "      <th>price_mid_peak_var</th>\n",
       "      <th>price_off_peak_fix</th>\n",
       "      <th>price_peak_fix</th>\n",
       "      <th>price_mid_peak_fix</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0002203ffbb812588b632b9e628cc38d</td>\n",
       "      <td>1.492061</td>\n",
       "      <td>1.245525</td>\n",
       "      <td>0.877924</td>\n",
       "      <td>488.420789</td>\n",
       "      <td>293.052462</td>\n",
       "      <td>195.368327</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0004351ebdd665e6ee664792efc4fd13</td>\n",
       "      <td>1.757118</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>532.625404</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0010bcc39e42b3c2131ed2ce55246e3c</td>\n",
       "      <td>2.178702</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>543.836520</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0010ee3855fdea87602a5b7aba8e42de</td>\n",
       "      <td>1.425085</td>\n",
       "      <td>1.179509</td>\n",
       "      <td>0.828384</td>\n",
       "      <td>487.769130</td>\n",
       "      <td>292.661462</td>\n",
       "      <td>195.107656</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>00114d74e963e47177db89bc70108537</td>\n",
       "      <td>1.775110</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>531.203164</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 id  price_off_peak_var  price_peak_var  \\\n",
       "0  0002203ffbb812588b632b9e628cc38d            1.492061        1.245525   \n",
       "1  0004351ebdd665e6ee664792efc4fd13            1.757118        0.000000   \n",
       "2  0010bcc39e42b3c2131ed2ce55246e3c            2.178702        0.000000   \n",
       "3  0010ee3855fdea87602a5b7aba8e42de            1.425085        1.179509   \n",
       "4  00114d74e963e47177db89bc70108537            1.775110        0.000000   \n",
       "\n",
       "   price_mid_peak_var  price_off_peak_fix  price_peak_fix  price_mid_peak_fix  \n",
       "0            0.877924          488.420789      293.052462          195.368327  \n",
       "1            0.000000          532.625404        0.000000            0.000000  \n",
       "2            0.000000          543.836520        0.000000            0.000000  \n",
       "3            0.828384          487.769130      292.661462          195.107656  \n",
       "4            0.000000          531.203164        0.000000            0.000000  "
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Lets group the data year wise and calculate sum for each customer.\n",
    "yr_df=dfp.groupby(['id'],as_index=False).sum()\n",
    "yr_df.drop('Month',1,inplace=True)\n",
    "yr_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "id": "136060b6",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((16096, 7), (16096, 2))"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "yr_df.shape,month_cnt.shape  #Equality of records.Merging will be lossless."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "id": "a38c1bde",
   "metadata": {},
   "outputs": [],
   "source": [
    "sum_df=pd.merge(yr_df,month_cnt,on='id')\n",
    "sum_df['Total_var']=sum_df.iloc[:,1:4].sum(axis=1)\n",
    "sum_df['Total_fix']=sum_df.iloc[:,4:7].sum(axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "id": "839f7f13",
   "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>id</th>\n",
       "      <th>price_off_peak_var</th>\n",
       "      <th>price_peak_var</th>\n",
       "      <th>price_mid_peak_var</th>\n",
       "      <th>price_off_peak_fix</th>\n",
       "      <th>price_peak_fix</th>\n",
       "      <th>price_mid_peak_fix</th>\n",
       "      <th>Month</th>\n",
       "      <th>Total_var</th>\n",
       "      <th>Total_fix</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0002203ffbb812588b632b9e628cc38d</td>\n",
       "      <td>1.492061</td>\n",
       "      <td>1.245525</td>\n",
       "      <td>0.877924</td>\n",
       "      <td>488.420789</td>\n",
       "      <td>293.052462</td>\n",
       "      <td>195.368327</td>\n",
       "      <td>12</td>\n",
       "      <td>3.615510</td>\n",
       "      <td>976.841578</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0004351ebdd665e6ee664792efc4fd13</td>\n",
       "      <td>1.757118</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>532.625404</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>12</td>\n",
       "      <td>1.757118</td>\n",
       "      <td>532.625404</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0010bcc39e42b3c2131ed2ce55246e3c</td>\n",
       "      <td>2.178702</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>543.836520</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>12</td>\n",
       "      <td>2.178702</td>\n",
       "      <td>543.836520</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>0010ee3855fdea87602a5b7aba8e42de</td>\n",
       "      <td>1.425085</td>\n",
       "      <td>1.179509</td>\n",
       "      <td>0.828384</td>\n",
       "      <td>487.769130</td>\n",
       "      <td>292.661462</td>\n",
       "      <td>195.107656</td>\n",
       "      <td>12</td>\n",
       "      <td>3.432978</td>\n",
       "      <td>975.538248</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>00114d74e963e47177db89bc70108537</td>\n",
       "      <td>1.775110</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>531.203164</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>12</td>\n",
       "      <td>1.775110</td>\n",
       "      <td>531.203164</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 id  price_off_peak_var  price_peak_var  \\\n",
       "0  0002203ffbb812588b632b9e628cc38d            1.492061        1.245525   \n",
       "1  0004351ebdd665e6ee664792efc4fd13            1.757118        0.000000   \n",
       "2  0010bcc39e42b3c2131ed2ce55246e3c            2.178702        0.000000   \n",
       "3  0010ee3855fdea87602a5b7aba8e42de            1.425085        1.179509   \n",
       "4  00114d74e963e47177db89bc70108537            1.775110        0.000000   \n",
       "\n",
       "   price_mid_peak_var  price_off_peak_fix  price_peak_fix  price_mid_peak_fix  \\\n",
       "0            0.877924          488.420789      293.052462          195.368327   \n",
       "1            0.000000          532.625404        0.000000            0.000000   \n",
       "2            0.000000          543.836520        0.000000            0.000000   \n",
       "3            0.828384          487.769130      292.661462          195.107656   \n",
       "4            0.000000          531.203164        0.000000            0.000000   \n",
       "\n",
       "   Month  Total_var   Total_fix  \n",
       "0     12   3.615510  976.841578  \n",
       "1     12   1.757118  532.625404  \n",
       "2     12   2.178702  543.836520  \n",
       "3     12   3.432978  975.538248  \n",
       "4     12   1.775110  531.203164  "
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sum_df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "id": "5fca9a17",
   "metadata": {},
   "outputs": [],
   "source": [
    "sum_df.rename(columns={'Month':'Month_Count'},inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "id": "eb3047ec",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "((14606, 26), (16096, 10))"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#Lets merge this with original dataframe\n",
    "\n",
    "dfc.shape,sum_df.shape\n",
    "\n",
    "#This shows there are certain company id present in price dataset(price_df) which are not present in client dataset(client_df)\n",
    "#Merging will result in data losses."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "8a049239",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(14606, 35)"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dft=pd.merge(dfc,sum_df,on='id')\n",
    "dft.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "f9b0e407",
   "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>id</th>\n",
       "      <th>channel_sales</th>\n",
       "      <th>cons_12m</th>\n",
       "      <th>cons_gas_12m</th>\n",
       "      <th>cons_last_month</th>\n",
       "      <th>date_activ</th>\n",
       "      <th>date_end</th>\n",
       "      <th>date_modif_prod</th>\n",
       "      <th>date_renewal</th>\n",
       "      <th>forecast_cons_12m</th>\n",
       "      <th>...</th>\n",
       "      <th>churn</th>\n",
       "      <th>price_off_peak_var</th>\n",
       "      <th>price_peak_var</th>\n",
       "      <th>price_mid_peak_var</th>\n",
       "      <th>price_off_peak_fix</th>\n",
       "      <th>price_peak_fix</th>\n",
       "      <th>price_mid_peak_fix</th>\n",
       "      <th>Month_Count</th>\n",
       "      <th>Total_var</th>\n",
       "      <th>Total_fix</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>24011ae4ebbe3035111d65fa7c15bc57</td>\n",
       "      <td>foosdfpfkusacimwkcsosbicdxkicaua</td>\n",
       "      <td>0</td>\n",
       "      <td>54946</td>\n",
       "      <td>0</td>\n",
       "      <td>2013-06-15</td>\n",
       "      <td>2016-06-15</td>\n",
       "      <td>2015-11-01</td>\n",
       "      <td>2015-06-23</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>1</td>\n",
       "      <td>1.497445</td>\n",
       "      <td>1.208992</td>\n",
       "      <td>0.798365</td>\n",
       "      <td>491.307186</td>\n",
       "      <td>268.224124</td>\n",
       "      <td>178.816084</td>\n",
       "      <td>12</td>\n",
       "      <td>3.504802</td>\n",
       "      <td>938.347394</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>d29c2c54acc38ff3c0614d0a653813dd</td>\n",
       "      <td>MISSING</td>\n",
       "      <td>4660</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2009-08-21</td>\n",
       "      <td>2016-08-30</td>\n",
       "      <td>2009-08-21</td>\n",
       "      <td>2015-08-31</td>\n",
       "      <td>189.95</td>\n",
       "      <td>...</td>\n",
       "      <td>0</td>\n",
       "      <td>1.795307</td>\n",
       "      <td>0.085483</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>531.736503</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>12</td>\n",
       "      <td>1.880790</td>\n",
       "      <td>531.736503</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>2 rows × 35 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 id                     channel_sales  \\\n",
       "0  24011ae4ebbe3035111d65fa7c15bc57  foosdfpfkusacimwkcsosbicdxkicaua   \n",
       "1  d29c2c54acc38ff3c0614d0a653813dd                           MISSING   \n",
       "\n",
       "   cons_12m  cons_gas_12m  cons_last_month date_activ   date_end  \\\n",
       "0         0         54946                0 2013-06-15 2016-06-15   \n",
       "1      4660             0                0 2009-08-21 2016-08-30   \n",
       "\n",
       "  date_modif_prod date_renewal  forecast_cons_12m  ...  churn  \\\n",
       "0      2015-11-01   2015-06-23               0.00  ...      1   \n",
       "1      2009-08-21   2015-08-31             189.95  ...      0   \n",
       "\n",
       "   price_off_peak_var  price_peak_var  price_mid_peak_var  price_off_peak_fix  \\\n",
       "0            1.497445        1.208992            0.798365          491.307186   \n",
       "1            1.795307        0.085483            0.000000          531.736503   \n",
       "\n",
       "   price_peak_fix price_mid_peak_fix  Month_Count  Total_var   Total_fix  \n",
       "0      268.224124         178.816084           12   3.504802  938.347394  \n",
       "1        0.000000           0.000000           12   1.880790  531.736503  \n",
       "\n",
       "[2 rows x 35 columns]"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dft.head(2)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "a246367e",
   "metadata": {},
   "source": [
    "## Corelation check for price paid by each customer for a year affecting churn"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "5826b33f",
   "metadata": {},
   "outputs": [],
   "source": [
    "# First step would be finding out all columns which define price of product.\n",
    "df=dft.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "3dbb9822",
   "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>churn</th>\n",
       "      <th>price_off_peak_var</th>\n",
       "      <th>price_peak_var</th>\n",
       "      <th>price_mid_peak_var</th>\n",
       "      <th>price_off_peak_fix</th>\n",
       "      <th>price_peak_fix</th>\n",
       "      <th>price_mid_peak_fix</th>\n",
       "      <th>Month_Count</th>\n",
       "      <th>Total_var</th>\n",
       "      <th>Total_fix</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1</td>\n",
       "      <td>1.497445</td>\n",
       "      <td>1.208992</td>\n",
       "      <td>0.798365</td>\n",
       "      <td>491.307186</td>\n",
       "      <td>268.224124</td>\n",
       "      <td>178.816084</td>\n",
       "      <td>12</td>\n",
       "      <td>3.504802</td>\n",
       "      <td>938.347394</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0</td>\n",
       "      <td>1.795307</td>\n",
       "      <td>0.085483</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>531.736503</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>12</td>\n",
       "      <td>1.880790</td>\n",
       "      <td>531.736503</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   churn  price_off_peak_var  price_peak_var  price_mid_peak_var  \\\n",
       "0      1            1.497445        1.208992            0.798365   \n",
       "1      0            1.795307        0.085483            0.000000   \n",
       "\n",
       "   price_off_peak_fix  price_peak_fix  price_mid_peak_fix  Month_Count  \\\n",
       "0          491.307186      268.224124          178.816084           12   \n",
       "1          531.736503        0.000000            0.000000           12   \n",
       "\n",
       "   Total_var   Total_fix  \n",
       "0   3.504802  938.347394  \n",
       "1   1.880790  531.736503  "
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "df_price=df.iloc[:,25:35]\n",
    "df_price.churn=df_price.churn.astype('int')\n",
    "df_price.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "id": "b01d08db",
   "metadata": {
    "scrolled": false
   },
   "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>churn</th>\n",
       "      <th>price_off_peak_var</th>\n",
       "      <th>price_peak_var</th>\n",
       "      <th>price_mid_peak_var</th>\n",
       "      <th>price_off_peak_fix</th>\n",
       "      <th>price_peak_fix</th>\n",
       "      <th>price_mid_peak_fix</th>\n",
       "      <th>Month_Count</th>\n",
       "      <th>Total_var</th>\n",
       "      <th>Total_fix</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>churn</th>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.007860</td>\n",
       "      <td>0.029280</td>\n",
       "      <td>0.046249</td>\n",
       "      <td>0.014435</td>\n",
       "      <td>0.046952</td>\n",
       "      <td>0.044533</td>\n",
       "      <td>-0.023343</td>\n",
       "      <td>0.039546</td>\n",
       "      <td>0.051520</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>price_off_peak_var</th>\n",
       "      <td>-0.007860</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.316327</td>\n",
       "      <td>-0.607230</td>\n",
       "      <td>0.664889</td>\n",
       "      <td>-0.629758</td>\n",
       "      <td>-0.607400</td>\n",
       "      <td>0.080187</td>\n",
       "      <td>-0.200873</td>\n",
       "      <td>-0.488141</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>price_peak_var</th>\n",
       "      <td>0.029280</td>\n",
       "      <td>-0.316327</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.822323</td>\n",
       "      <td>-0.209816</td>\n",
       "      <td>0.810536</td>\n",
       "      <td>0.815556</td>\n",
       "      <td>0.002376</td>\n",
       "      <td>0.971263</td>\n",
       "      <td>0.797857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>price_mid_peak_var</th>\n",
       "      <td>0.046249</td>\n",
       "      <td>-0.607230</td>\n",
       "      <td>0.822323</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.286856</td>\n",
       "      <td>0.987688</td>\n",
       "      <td>0.991539</td>\n",
       "      <td>-0.005015</td>\n",
       "      <td>0.849157</td>\n",
       "      <td>0.963791</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>price_off_peak_fix</th>\n",
       "      <td>0.014435</td>\n",
       "      <td>0.664889</td>\n",
       "      <td>-0.209816</td>\n",
       "      <td>-0.286856</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.227803</td>\n",
       "      <td>-0.331147</td>\n",
       "      <td>0.103558</td>\n",
       "      <td>-0.076908</td>\n",
       "      <td>-0.039050</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>price_peak_fix</th>\n",
       "      <td>0.046952</td>\n",
       "      <td>-0.629758</td>\n",
       "      <td>0.810536</td>\n",
       "      <td>0.987688</td>\n",
       "      <td>-0.227803</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.974097</td>\n",
       "      <td>-0.005546</td>\n",
       "      <td>0.828465</td>\n",
       "      <td>0.978708</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>price_mid_peak_fix</th>\n",
       "      <td>0.044533</td>\n",
       "      <td>-0.607400</td>\n",
       "      <td>0.815556</td>\n",
       "      <td>0.991539</td>\n",
       "      <td>-0.331147</td>\n",
       "      <td>0.974097</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-0.005685</td>\n",
       "      <td>0.840486</td>\n",
       "      <td>0.947945</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Month_Count</th>\n",
       "      <td>-0.023343</td>\n",
       "      <td>0.080187</td>\n",
       "      <td>0.002376</td>\n",
       "      <td>-0.005015</td>\n",
       "      <td>0.103558</td>\n",
       "      <td>-0.005546</td>\n",
       "      <td>-0.005685</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.023556</td>\n",
       "      <td>0.019123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Total_var</th>\n",
       "      <td>0.039546</td>\n",
       "      <td>-0.200873</td>\n",
       "      <td>0.971263</td>\n",
       "      <td>0.849157</td>\n",
       "      <td>-0.076908</td>\n",
       "      <td>0.828465</td>\n",
       "      <td>0.840486</td>\n",
       "      <td>0.023556</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.851493</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Total_fix</th>\n",
       "      <td>0.051520</td>\n",
       "      <td>-0.488141</td>\n",
       "      <td>0.797857</td>\n",
       "      <td>0.963791</td>\n",
       "      <td>-0.039050</td>\n",
       "      <td>0.978708</td>\n",
       "      <td>0.947945</td>\n",
       "      <td>0.019123</td>\n",
       "      <td>0.851493</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                       churn  price_off_peak_var  price_peak_var  \\\n",
       "churn               1.000000           -0.007860        0.029280   \n",
       "price_off_peak_var -0.007860            1.000000       -0.316327   \n",
       "price_peak_var      0.029280           -0.316327        1.000000   \n",
       "price_mid_peak_var  0.046249           -0.607230        0.822323   \n",
       "price_off_peak_fix  0.014435            0.664889       -0.209816   \n",
       "price_peak_fix      0.046952           -0.629758        0.810536   \n",
       "price_mid_peak_fix  0.044533           -0.607400        0.815556   \n",
       "Month_Count        -0.023343            0.080187        0.002376   \n",
       "Total_var           0.039546           -0.200873        0.971263   \n",
       "Total_fix           0.051520           -0.488141        0.797857   \n",
       "\n",
       "                    price_mid_peak_var  price_off_peak_fix  price_peak_fix  \\\n",
       "churn                         0.046249            0.014435        0.046952   \n",
       "price_off_peak_var           -0.607230            0.664889       -0.629758   \n",
       "price_peak_var                0.822323           -0.209816        0.810536   \n",
       "price_mid_peak_var            1.000000           -0.286856        0.987688   \n",
       "price_off_peak_fix           -0.286856            1.000000       -0.227803   \n",
       "price_peak_fix                0.987688           -0.227803        1.000000   \n",
       "price_mid_peak_fix            0.991539           -0.331147        0.974097   \n",
       "Month_Count                  -0.005015            0.103558       -0.005546   \n",
       "Total_var                     0.849157           -0.076908        0.828465   \n",
       "Total_fix                     0.963791           -0.039050        0.978708   \n",
       "\n",
       "                    price_mid_peak_fix  Month_Count  Total_var  Total_fix  \n",
       "churn                         0.044533    -0.023343   0.039546   0.051520  \n",
       "price_off_peak_var           -0.607400     0.080187  -0.200873  -0.488141  \n",
       "price_peak_var                0.815556     0.002376   0.971263   0.797857  \n",
       "price_mid_peak_var            0.991539    -0.005015   0.849157   0.963791  \n",
       "price_off_peak_fix           -0.331147     0.103558  -0.076908  -0.039050  \n",
       "price_peak_fix                0.974097    -0.005546   0.828465   0.978708  \n",
       "price_mid_peak_fix            1.000000    -0.005685   0.840486   0.947945  \n",
       "Month_Count                  -0.005685     1.000000   0.023556   0.019123  \n",
       "Total_var                     0.840486     0.023556   1.000000   0.851493  \n",
       "Total_fix                     0.947945     0.019123   0.851493   1.000000  "
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_price.corr()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "11a81a5a",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "651af037",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "id": "02be2228",
   "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>id</th>\n",
       "      <th>price_off_peak_var</th>\n",
       "      <th>price_peak_var</th>\n",
       "      <th>price_mid_peak_var</th>\n",
       "      <th>price_off_peak_fix</th>\n",
       "      <th>price_peak_fix</th>\n",
       "      <th>price_mid_peak_fix</th>\n",
       "      <th>Month</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>-1.150185</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-0.000003</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>0.464491</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.401609</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>-2.968314</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>8</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>038af19179925da21a25619c5a24b745</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>31f2ce549924679a3cbb2d128ae9ea43</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                  id  price_off_peak_var  price_peak_var  \\\n",
       "0   038af19179925da21a25619c5a24b745                 NaN             NaN   \n",
       "1   038af19179925da21a25619c5a24b745            0.000000             NaN   \n",
       "2   038af19179925da21a25619c5a24b745            0.000000             NaN   \n",
       "3   038af19179925da21a25619c5a24b745           -1.150185             NaN   \n",
       "4   038af19179925da21a25619c5a24b745            0.000000             NaN   \n",
       "5   038af19179925da21a25619c5a24b745            0.000000             NaN   \n",
       "6   038af19179925da21a25619c5a24b745            0.464491             NaN   \n",
       "7   038af19179925da21a25619c5a24b745           -2.968314             NaN   \n",
       "8   038af19179925da21a25619c5a24b745            0.000000             NaN   \n",
       "9   038af19179925da21a25619c5a24b745            0.000000             NaN   \n",
       "10  038af19179925da21a25619c5a24b745            0.000000             NaN   \n",
       "11  038af19179925da21a25619c5a24b745            0.000000             NaN   \n",
       "12  31f2ce549924679a3cbb2d128ae9ea43                 NaN             NaN   \n",
       "\n",
       "    price_mid_peak_var  price_off_peak_fix  price_peak_fix  \\\n",
       "0                  NaN                 NaN             NaN   \n",
       "1                  NaN            0.000000             NaN   \n",
       "2                  NaN            0.000000             NaN   \n",
       "3                  NaN            0.000000             NaN   \n",
       "4                  NaN            0.000000             NaN   \n",
       "5                  NaN           -0.000003             NaN   \n",
       "6                  NaN            0.401609             NaN   \n",
       "7                  NaN            0.000000             NaN   \n",
       "8                  NaN            0.000000             NaN   \n",
       "9                  NaN            0.000000             NaN   \n",
       "10                 NaN            0.000000             NaN   \n",
       "11                 NaN            0.000000             NaN   \n",
       "12                 NaN                 NaN             NaN   \n",
       "\n",
       "    price_mid_peak_fix  Month  \n",
       "0                  NaN      1  \n",
       "1                  NaN      2  \n",
       "2                  NaN      3  \n",
       "3                  NaN      4  \n",
       "4                  NaN      5  \n",
       "5                  NaN      6  \n",
       "6                  NaN      7  \n",
       "7                  NaN      8  \n",
       "8                  NaN      9  \n",
       "9                  NaN     10  \n",
       "10                 NaN     11  \n",
       "11                 NaN     12  \n",
       "12                 NaN      1  "
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "price_sen=pd.read_csv('Price_Sensitivity.csv')\n",
    "price_sen.head(13)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "id": "8be8210a",
   "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>id</th>\n",
       "      <th>price_off_peak_var</th>\n",
       "      <th>price_peak_var</th>\n",
       "      <th>price_mid_peak_var</th>\n",
       "      <th>price_off_peak_fix</th>\n",
       "      <th>price_peak_fix</th>\n",
       "      <th>price_mid_peak_fix</th>\n",
       "      <th>Month</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>11231</th>\n",
       "      <td>17bf5be2125f6e81b22e9121f58ad873</td>\n",
       "      <td>inf</td>\n",
       "      <td>inf</td>\n",
       "      <td>inf</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16813</th>\n",
       "      <td>a8909ca0c4a85f075fc91b55a68cd11c</td>\n",
       "      <td>inf</td>\n",
       "      <td>inf</td>\n",
       "      <td>inf</td>\n",
       "      <td>inf</td>\n",
       "      <td>inf</td>\n",
       "      <td>inf</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18625</th>\n",
       "      <td>793577ee238c6d5dac50c4d968e02421</td>\n",
       "      <td>inf</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>inf</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>12</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27664</th>\n",
       "      <td>2f527996b16726de6db3f2b5b33746e2</td>\n",
       "      <td>inf</td>\n",
       "      <td>inf</td>\n",
       "      <td>NaN</td>\n",
       "      <td>inf</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>11</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30012</th>\n",
       "      <td>c89261a0d934d2257420fff7292db0c0</td>\n",
       "      <td>inf</td>\n",
       "      <td>inf</td>\n",
       "      <td>inf</td>\n",
       "      <td>inf</td>\n",
       "      <td>inf</td>\n",
       "      <td>inf</td>\n",
       "      <td>10</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                     id  price_off_peak_var  price_peak_var  \\\n",
       "11231  17bf5be2125f6e81b22e9121f58ad873                 inf             inf   \n",
       "16813  a8909ca0c4a85f075fc91b55a68cd11c                 inf             inf   \n",
       "18625  793577ee238c6d5dac50c4d968e02421                 inf             NaN   \n",
       "27664  2f527996b16726de6db3f2b5b33746e2                 inf             inf   \n",
       "30012  c89261a0d934d2257420fff7292db0c0                 inf             inf   \n",
       "\n",
       "       price_mid_peak_var  price_off_peak_fix  price_peak_fix  \\\n",
       "11231                 inf                 0.0             0.0   \n",
       "16813                 inf                 inf             inf   \n",
       "18625                 NaN                 inf             NaN   \n",
       "27664                 NaN                 inf             NaN   \n",
       "30012                 inf                 inf             inf   \n",
       "\n",
       "       price_mid_peak_fix  Month  \n",
       "11231                 0.0      4  \n",
       "16813                 inf     11  \n",
       "18625                 NaN     12  \n",
       "27664                 NaN     11  \n",
       "30012                 inf     10  "
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "price_sen[price_sen.price_off_peak_var==np.inf].head()\n",
    "#These are the values for which previous price is 0, after price is 0 , so its giving a Infinite value\n",
    "#So it comes under value greater than 0 which van be misleading.\n",
    "#Lets replace them with 0 as 0 means no change in price sensitivity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "887228d9",
   "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>id</th>\n",
       "      <th>price_off_peak_var</th>\n",
       "      <th>price_peak_var</th>\n",
       "      <th>price_mid_peak_var</th>\n",
       "      <th>price_off_peak_fix</th>\n",
       "      <th>price_peak_fix</th>\n",
       "      <th>price_mid_peak_fix</th>\n",
       "      <th>Month</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "Empty DataFrame\n",
       "Columns: [id, price_off_peak_var, price_peak_var, price_mid_peak_var, price_off_peak_fix, price_peak_fix, price_mid_peak_fix, Month]\n",
       "Index: []"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "price_sen.replace({np.inf:0},inplace=True)\n",
    "price_sen[price_sen.price_off_peak_var==np.inf].head()\n",
    "#They are removed from our dataframe."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "id": "b674364a",
   "metadata": {},
   "outputs": [],
   "source": [
    "cols=price_sen.columns[1:7]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "id": "7c4957cd",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Lets define various derived parameters from price sensitvity\n",
    "\n",
    "#price_sen_negative_values_count=count of all negative price sensitivity for a given customer\n",
    "#price_sen_negative_value_sum=sum of all negative price sensitivity for a given customer\n",
    "#Price_sen_negative_value_avg=price_sen_negative_value_sum/price_sen_negative_values_count\n",
    "\n",
    "#Similar for postive price sensitivity\n",
    "\n",
    "#price_sen_negative_values_max=maximum negative price sensitivity(least negative value)\n",
    "#price_sen_positive_values_max=maximul positive price sensitivity (largest positive value.)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "id": "6a3bd1f2",
   "metadata": {},
   "outputs": [],
   "source": [
    "price_sen_negative_values_count=price_sen.groupby('id')[cols].apply(lambda x:x[x<0].count())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "id": "3c85d3d7",
   "metadata": {},
   "outputs": [],
   "source": [
    "price_sen_negative_values_sum=price_sen.groupby('id')[cols].apply(lambda x:x[x<0].sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "id": "507fd426",
   "metadata": {},
   "outputs": [],
   "source": [
    "price_sen_positive_values_count=price_sen.groupby('id')[cols].apply(lambda x:x[x>0].count())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "id": "925e6270",
   "metadata": {},
   "outputs": [],
   "source": [
    "price_sen_positive_values_sum=price_sen.groupby('id')[cols].apply(lambda x:x[x>0].sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "id": "7e4616b1",
   "metadata": {},
   "outputs": [],
   "source": [
    "price_sen_negative_value_avg=price_sen_negative_values_sum/price_sen_negative_values_count\n",
    "price_sen_positive_value_avg=price_sen_positive_values_sum/price_sen_positive_values_count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "id": "280fa7f8",
   "metadata": {},
   "outputs": [],
   "source": [
    "price_sen_negative_value_max=price_sen.groupby('id')[cols].apply(lambda x:x[x<0].min())\n",
    "price_sen_positive_value_max=price_sen.groupby('id')[cols].apply(lambda x:x[x>0].max())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "id": "4086c74b",
   "metadata": {},
   "outputs": [],
   "source": [
    "target=dft[['id','churn']] #Creation churn as our target variable.\n",
    "target.churn=target.churn.astype('int')"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "6fa74755",
   "metadata": {},
   "source": [
    "## Corelation for various parameters obtained above"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "id": "455eee1f",
   "metadata": {},
   "outputs": [],
   "source": [
    "def cdata(ps_df,y=target):\n",
    "    ps_df=pd.merge(ps_df,target,on='id')\n",
    "    x=ps_df.corr().iloc[:,-1]\n",
    "    return x"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "id": "b622fbe7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "price_off_peak_var   -0.006521\n",
       "price_peak_var        0.028554\n",
       "price_mid_peak_var   -0.000535\n",
       "price_off_peak_fix    0.019123\n",
       "price_peak_fix       -0.007635\n",
       "price_mid_peak_fix   -0.010646\n",
       "churn                 1.000000\n",
       "Name: churn, dtype: float64"
      ]
     },
     "execution_count": 58,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cdata(price_sen_positive_value_avg)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "id": "e64dccff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "price_off_peak_var   -0.030125\n",
       "price_peak_var        0.004715\n",
       "price_mid_peak_var   -0.012413\n",
       "price_off_peak_fix   -0.027530\n",
       "price_peak_fix       -0.006861\n",
       "price_mid_peak_fix   -0.010058\n",
       "churn                 1.000000\n",
       "Name: churn, dtype: float64"
      ]
     },
     "execution_count": 59,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cdata(price_sen_negative_value_avg)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "id": "fec382e2",
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "price_off_peak_var   -0.006365\n",
       "price_peak_var        0.029164\n",
       "price_mid_peak_var    0.003336\n",
       "price_off_peak_fix    0.022482\n",
       "price_peak_fix        0.001261\n",
       "price_mid_peak_fix   -0.004389\n",
       "churn                 1.000000\n",
       "Name: churn, dtype: float64"
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cdata(price_sen_positive_value_max)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "id": "43eff79e",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "price_off_peak_var   -0.025631\n",
       "price_peak_var        0.001762\n",
       "price_mid_peak_var   -0.013134\n",
       "price_off_peak_fix   -0.026472\n",
       "price_peak_fix       -0.010094\n",
       "price_mid_peak_fix   -0.012849\n",
       "churn                 1.000000\n",
       "Name: churn, dtype: float64"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "cdata(price_sen_negative_value_max)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "c08d413e",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "id": "14956741",
   "metadata": {},
   "outputs": [],
   "source": [
    "import statsmodels.tsa.api as smt"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "id": "d3d67b62",
   "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>price_off_peak_var</th>\n",
       "      <th>price_peak_var</th>\n",
       "      <th>price_mid_peak_var</th>\n",
       "      <th>price_off_peak_fix</th>\n",
       "      <th>price_peak_fix</th>\n",
       "      <th>price_mid_peak_fix</th>\n",
       "      <th>Month</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>price_date</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2015-01-01</th>\n",
       "      <td>0.142568</td>\n",
       "      <td>0.054950</td>\n",
       "      <td>0.030333</td>\n",
       "      <td>43.226643</td>\n",
       "      <td>10.692921</td>\n",
       "      <td>6.455861</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2015-02-01</th>\n",
       "      <td>0.142767</td>\n",
       "      <td>0.055053</td>\n",
       "      <td>0.030367</td>\n",
       "      <td>43.238161</td>\n",
       "      <td>10.673719</td>\n",
       "      <td>6.449016</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            price_off_peak_var  price_peak_var  price_mid_peak_var  \\\n",
       "price_date                                                           \n",
       "2015-01-01            0.142568        0.054950            0.030333   \n",
       "2015-02-01            0.142767        0.055053            0.030367   \n",
       "\n",
       "            price_off_peak_fix  price_peak_fix  price_mid_peak_fix  Month  \n",
       "price_date                                                                 \n",
       "2015-01-01           43.226643       10.692921            6.455861      1  \n",
       "2015-02-01           43.238161       10.673719            6.449016      2  "
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_date=dfp.groupby('price_date').mean()\n",
    "df_date.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "id": "fcf5c48d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "execution_count": 64,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "smt.graphics.plot_acf(df_date.iloc[:,0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "id": "c64ff566",
   "metadata": {},
   "outputs": [],
   "source": [
    "# No proper patterns found."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "id": "a7e8dea0",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Lets find a difference between first and last month divided by first month to get price sensitivity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "id": "b1d488d9",
   "metadata": {},
   "outputs": [],
   "source": [
    "first_value=dfp.groupby('id').first()\n",
    "last_value=dfp.groupby('id').last()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "id": "069ddac8",
   "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>price_date</th>\n",
       "      <th>price_off_peak_var</th>\n",
       "      <th>price_peak_var</th>\n",
       "      <th>price_mid_peak_var</th>\n",
       "      <th>price_off_peak_fix</th>\n",
       "      <th>price_peak_fix</th>\n",
       "      <th>price_mid_peak_fix</th>\n",
       "      <th>Month</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>id</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0002203ffbb812588b632b9e628cc38d</th>\n",
       "      <td>2015-01-01</td>\n",
       "      <td>0.126098</td>\n",
       "      <td>0.103975</td>\n",
       "      <td>0.070232</td>\n",
       "      <td>40.565969</td>\n",
       "      <td>24.339581</td>\n",
       "      <td>16.226389</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0004351ebdd665e6ee664792efc4fd13</th>\n",
       "      <td>2015-01-01</td>\n",
       "      <td>0.148047</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>44.266931</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0010bcc39e42b3c2131ed2ce55246e3c</th>\n",
       "      <td>2015-01-01</td>\n",
       "      <td>0.150837</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>44.444710</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0010ee3855fdea87602a5b7aba8e42de</th>\n",
       "      <td>2015-01-01</td>\n",
       "      <td>0.123086</td>\n",
       "      <td>0.100505</td>\n",
       "      <td>0.068646</td>\n",
       "      <td>40.565969</td>\n",
       "      <td>24.339581</td>\n",
       "      <td>16.226389</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>00114d74e963e47177db89bc70108537</th>\n",
       "      <td>2015-01-01</td>\n",
       "      <td>0.149434</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>44.266931</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 price_date  price_off_peak_var  \\\n",
       "id                                                                \n",
       "0002203ffbb812588b632b9e628cc38d 2015-01-01            0.126098   \n",
       "0004351ebdd665e6ee664792efc4fd13 2015-01-01            0.148047   \n",
       "0010bcc39e42b3c2131ed2ce55246e3c 2015-01-01            0.150837   \n",
       "0010ee3855fdea87602a5b7aba8e42de 2015-01-01            0.123086   \n",
       "00114d74e963e47177db89bc70108537 2015-01-01            0.149434   \n",
       "\n",
       "                                  price_peak_var  price_mid_peak_var  \\\n",
       "id                                                                     \n",
       "0002203ffbb812588b632b9e628cc38d        0.103975            0.070232   \n",
       "0004351ebdd665e6ee664792efc4fd13        0.000000            0.000000   \n",
       "0010bcc39e42b3c2131ed2ce55246e3c        0.000000            0.000000   \n",
       "0010ee3855fdea87602a5b7aba8e42de        0.100505            0.068646   \n",
       "00114d74e963e47177db89bc70108537        0.000000            0.000000   \n",
       "\n",
       "                                  price_off_peak_fix  price_peak_fix  \\\n",
       "id                                                                     \n",
       "0002203ffbb812588b632b9e628cc38d           40.565969       24.339581   \n",
       "0004351ebdd665e6ee664792efc4fd13           44.266931        0.000000   \n",
       "0010bcc39e42b3c2131ed2ce55246e3c           44.444710        0.000000   \n",
       "0010ee3855fdea87602a5b7aba8e42de           40.565969       24.339581   \n",
       "00114d74e963e47177db89bc70108537           44.266931        0.000000   \n",
       "\n",
       "                                  price_mid_peak_fix  Month  \n",
       "id                                                           \n",
       "0002203ffbb812588b632b9e628cc38d           16.226389      1  \n",
       "0004351ebdd665e6ee664792efc4fd13            0.000000      1  \n",
       "0010bcc39e42b3c2131ed2ce55246e3c            0.000000      1  \n",
       "0010ee3855fdea87602a5b7aba8e42de           16.226389      1  \n",
       "00114d74e963e47177db89bc70108537            0.000000      1  "
      ]
     },
     "execution_count": 68,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "first_value.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "id": "cdd21b19",
   "metadata": {},
   "outputs": [],
   "source": [
    "price_sen_year=(last_value.iloc[:,1:7]-first_value.iloc[:,1:7])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "id": "a7a4ebc5",
   "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>id</th>\n",
       "      <th>price_off_peak_var</th>\n",
       "      <th>price_peak_var</th>\n",
       "      <th>price_mid_peak_var</th>\n",
       "      <th>price_off_peak_fix</th>\n",
       "      <th>price_peak_fix</th>\n",
       "      <th>price_mid_peak_fix</th>\n",
       "      <th>churn</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0002203ffbb812588b632b9e628cc38d</td>\n",
       "      <td>-0.006192</td>\n",
       "      <td>-0.002302</td>\n",
       "      <td>0.003487</td>\n",
       "      <td>0.162916</td>\n",
       "      <td>0.097749</td>\n",
       "      <td>0.065166</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>0004351ebdd665e6ee664792efc4fd13</td>\n",
       "      <td>-0.004104</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.177779</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 id  price_off_peak_var  price_peak_var  \\\n",
       "0  0002203ffbb812588b632b9e628cc38d           -0.006192       -0.002302   \n",
       "1  0004351ebdd665e6ee664792efc4fd13           -0.004104        0.000000   \n",
       "\n",
       "   price_mid_peak_var  price_off_peak_fix  price_peak_fix  price_mid_peak_fix  \\\n",
       "0            0.003487            0.162916        0.097749            0.065166   \n",
       "1            0.000000            0.177779        0.000000            0.000000   \n",
       "\n",
       "   churn  \n",
       "0      0  \n",
       "1      0  "
      ]
     },
     "execution_count": 70,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "price_sen_year=pd.merge(price_sen_year,target,on='id')\n",
    "price_sen_year.head(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "id": "92276aff",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "price_off_peak_var   -0.001064\n",
       "price_peak_var       -0.004041\n",
       "price_mid_peak_var    0.000191\n",
       "price_off_peak_fix    0.002672\n",
       "price_peak_fix       -0.005183\n",
       "price_mid_peak_fix   -0.002681\n",
       "churn                 1.000000\n",
       "Name: churn, dtype: float64"
      ]
     },
     "execution_count": 71,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "price_sen_year.corr().iloc[:,-1]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "id": "90bd607d",
   "metadata": {},
   "outputs": [],
   "source": [
    "# We found price_sen_negative_values_avg to have maximum correlation with churn for price_off_peak_var and price_off_peak_fix\n",
    "#Similar pattern is obserbed for off-peak values.\n",
    "#Also in our dft dataframe we have defined groupby sum of every price for every customer, they all have comparable correlation.\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "id": "b20f43c0",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Lets aggregate all the columns having maximum correlation with churn.\n",
    "\n",
    "c1=price_sen_negative_value_avg.iloc[:,[0,3]]\n",
    "c1.columns=['off_peak_energy_negative_avg','off_peak_power_negative_avg']\n",
    "\n",
    "c2=price_sen_negative_value_max.iloc[:,[0,3]]\n",
    "c2.columns=['off_peak_energy_negative_max','off_peak_power_negative_max']\n",
    "\n",
    "\n",
    "\n",
    "c3=price_sen_positive_value_avg.iloc[:,[0,3]]\n",
    "c3.columns=['off_peak_energy_positive_avg','off_peak_power_positive_avg']\n",
    "\n",
    "\n",
    "c4=price_sen_positive_value_max.iloc[:,[0,3]]\n",
    "c4.columns=['off_peak_energy_positive_max','off_peak_power_positive_max']\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "id": "ce6751d9",
   "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>id</th>\n",
       "      <th>channel_sales</th>\n",
       "      <th>cons_12m</th>\n",
       "      <th>cons_gas_12m</th>\n",
       "      <th>cons_last_month</th>\n",
       "      <th>date_activ</th>\n",
       "      <th>date_end</th>\n",
       "      <th>date_modif_prod</th>\n",
       "      <th>date_renewal</th>\n",
       "      <th>forecast_cons_12m</th>\n",
       "      <th>...</th>\n",
       "      <th>Total_var</th>\n",
       "      <th>Total_fix</th>\n",
       "      <th>off_peak_energy_negative_avg</th>\n",
       "      <th>off_peak_power_negative_avg</th>\n",
       "      <th>off_peak_energy_negative_max</th>\n",
       "      <th>off_peak_power_negative_max</th>\n",
       "      <th>off_peak_energy_positive_avg</th>\n",
       "      <th>off_peak_power_positive_avg</th>\n",
       "      <th>off_peak_energy_positive_max</th>\n",
       "      <th>off_peak_power_positive_max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>24011ae4ebbe3035111d65fa7c15bc57</td>\n",
       "      <td>foosdfpfkusacimwkcsosbicdxkicaua</td>\n",
       "      <td>0</td>\n",
       "      <td>54946</td>\n",
       "      <td>0</td>\n",
       "      <td>2013-06-15</td>\n",
       "      <td>2016-06-15</td>\n",
       "      <td>2015-11-01</td>\n",
       "      <td>2015-06-23</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>3.504802</td>\n",
       "      <td>938.347394</td>\n",
       "      <td>-3.381130</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-6.495543</td>\n",
       "      <td>NaN</td>\n",
       "      <td>24.305621</td>\n",
       "      <td>3.029476</td>\n",
       "      <td>24.305621</td>\n",
       "      <td>8.686820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>d29c2c54acc38ff3c0614d0a653813dd</td>\n",
       "      <td>MISSING</td>\n",
       "      <td>4660</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2009-08-21</td>\n",
       "      <td>2016-08-30</td>\n",
       "      <td>2009-08-21</td>\n",
       "      <td>2015-08-31</td>\n",
       "      <td>189.95</td>\n",
       "      <td>...</td>\n",
       "      <td>1.880790</td>\n",
       "      <td>531.736503</td>\n",
       "      <td>-1.770692</td>\n",
       "      <td>-0.000003</td>\n",
       "      <td>-2.947802</td>\n",
       "      <td>-0.000003</td>\n",
       "      <td>1.073045</td>\n",
       "      <td>0.401609</td>\n",
       "      <td>1.073045</td>\n",
       "      <td>0.401609</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>764c75f661154dac3a6c254cd082ea7d</td>\n",
       "      <td>foosdfpfkusacimwkcsosbicdxkicaua</td>\n",
       "      <td>544</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2010-04-16</td>\n",
       "      <td>2016-04-16</td>\n",
       "      <td>2010-04-16</td>\n",
       "      <td>2015-04-17</td>\n",
       "      <td>47.96</td>\n",
       "      <td>...</td>\n",
       "      <td>3.107195</td>\n",
       "      <td>532.625404</td>\n",
       "      <td>-1.354201</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-2.684050</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.401606</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.401606</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>bba03439a292a1e166f80264c16191cb</td>\n",
       "      <td>lmkebamcaaclubfxadlmueccxoimlema</td>\n",
       "      <td>1584</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2010-03-30</td>\n",
       "      <td>2016-03-30</td>\n",
       "      <td>2010-03-30</td>\n",
       "      <td>2015-03-31</td>\n",
       "      <td>240.04</td>\n",
       "      <td>...</td>\n",
       "      <td>1.814521</td>\n",
       "      <td>532.803183</td>\n",
       "      <td>-1.485466</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-2.915425</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.401606</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.401606</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>149d57cf92fc41cf94415803a877cb4b</td>\n",
       "      <td>MISSING</td>\n",
       "      <td>4425</td>\n",
       "      <td>0</td>\n",
       "      <td>526</td>\n",
       "      <td>2010-01-13</td>\n",
       "      <td>2016-03-07</td>\n",
       "      <td>2010-01-13</td>\n",
       "      <td>2015-03-09</td>\n",
       "      <td>445.75</td>\n",
       "      <td>...</td>\n",
       "      <td>3.608133</td>\n",
       "      <td>976.515746</td>\n",
       "      <td>-6.372446</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-6.372446</td>\n",
       "      <td>NaN</td>\n",
       "      <td>1.561484</td>\n",
       "      <td>0.401607</td>\n",
       "      <td>1.561484</td>\n",
       "      <td>0.401607</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 43 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 id                     channel_sales  \\\n",
       "0  24011ae4ebbe3035111d65fa7c15bc57  foosdfpfkusacimwkcsosbicdxkicaua   \n",
       "1  d29c2c54acc38ff3c0614d0a653813dd                           MISSING   \n",
       "2  764c75f661154dac3a6c254cd082ea7d  foosdfpfkusacimwkcsosbicdxkicaua   \n",
       "3  bba03439a292a1e166f80264c16191cb  lmkebamcaaclubfxadlmueccxoimlema   \n",
       "4  149d57cf92fc41cf94415803a877cb4b                           MISSING   \n",
       "\n",
       "   cons_12m  cons_gas_12m  cons_last_month date_activ   date_end  \\\n",
       "0         0         54946                0 2013-06-15 2016-06-15   \n",
       "1      4660             0                0 2009-08-21 2016-08-30   \n",
       "2       544             0                0 2010-04-16 2016-04-16   \n",
       "3      1584             0                0 2010-03-30 2016-03-30   \n",
       "4      4425             0              526 2010-01-13 2016-03-07   \n",
       "\n",
       "  date_modif_prod date_renewal  forecast_cons_12m  ...  Total_var   Total_fix  \\\n",
       "0      2015-11-01   2015-06-23               0.00  ...   3.504802  938.347394   \n",
       "1      2009-08-21   2015-08-31             189.95  ...   1.880790  531.736503   \n",
       "2      2010-04-16   2015-04-17              47.96  ...   3.107195  532.625404   \n",
       "3      2010-03-30   2015-03-31             240.04  ...   1.814521  532.803183   \n",
       "4      2010-01-13   2015-03-09             445.75  ...   3.608133  976.515746   \n",
       "\n",
       "   off_peak_energy_negative_avg  off_peak_power_negative_avg  \\\n",
       "0                     -3.381130                          NaN   \n",
       "1                     -1.770692                    -0.000003   \n",
       "2                     -1.354201                          NaN   \n",
       "3                     -1.485466                          NaN   \n",
       "4                     -6.372446                          NaN   \n",
       "\n",
       "   off_peak_energy_negative_max  off_peak_power_negative_max  \\\n",
       "0                     -6.495543                          NaN   \n",
       "1                     -2.947802                    -0.000003   \n",
       "2                     -2.684050                          NaN   \n",
       "3                     -2.915425                          NaN   \n",
       "4                     -6.372446                          NaN   \n",
       "\n",
       "  off_peak_energy_positive_avg  off_peak_power_positive_avg  \\\n",
       "0                    24.305621                     3.029476   \n",
       "1                     1.073045                     0.401609   \n",
       "2                          NaN                     0.401606   \n",
       "3                          NaN                     0.401606   \n",
       "4                     1.561484                     0.401607   \n",
       "\n",
       "   off_peak_energy_positive_max  off_peak_power_positive_max  \n",
       "0                     24.305621                     8.686820  \n",
       "1                      1.073045                     0.401609  \n",
       "2                           NaN                     0.401606  \n",
       "3                           NaN                     0.401606  \n",
       "4                      1.561484                     0.401607  \n",
       "\n",
       "[5 rows x 43 columns]"
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dft=pd.merge(dft,c1,on='id')\n",
    "dft=pd.merge(dft,c2,on='id')\n",
    "dft=pd.merge(dft,c3,on='id')\n",
    "dft=pd.merge(dft,c4,on='id')\n",
    "dft.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "id": "e76eaf3b",
   "metadata": {},
   "outputs": [],
   "source": [
    "dft.replace(np.nan,0,inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "id": "e576e1d8",
   "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>id</th>\n",
       "      <th>channel_sales</th>\n",
       "      <th>cons_12m</th>\n",
       "      <th>cons_gas_12m</th>\n",
       "      <th>cons_last_month</th>\n",
       "      <th>date_activ</th>\n",
       "      <th>date_end</th>\n",
       "      <th>date_modif_prod</th>\n",
       "      <th>date_renewal</th>\n",
       "      <th>forecast_cons_12m</th>\n",
       "      <th>...</th>\n",
       "      <th>Total_var</th>\n",
       "      <th>Total_fix</th>\n",
       "      <th>off_peak_energy_negative_avg</th>\n",
       "      <th>off_peak_power_negative_avg</th>\n",
       "      <th>off_peak_energy_negative_max</th>\n",
       "      <th>off_peak_power_negative_max</th>\n",
       "      <th>off_peak_energy_positive_avg</th>\n",
       "      <th>off_peak_power_positive_avg</th>\n",
       "      <th>off_peak_energy_positive_max</th>\n",
       "      <th>off_peak_power_positive_max</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>24011ae4ebbe3035111d65fa7c15bc57</td>\n",
       "      <td>foosdfpfkusacimwkcsosbicdxkicaua</td>\n",
       "      <td>0</td>\n",
       "      <td>54946</td>\n",
       "      <td>0</td>\n",
       "      <td>2013-06-15</td>\n",
       "      <td>2016-06-15</td>\n",
       "      <td>2015-11-01</td>\n",
       "      <td>2015-06-23</td>\n",
       "      <td>0.00</td>\n",
       "      <td>...</td>\n",
       "      <td>3.504802</td>\n",
       "      <td>938.347394</td>\n",
       "      <td>-3.381130</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-6.495543</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>24.305621</td>\n",
       "      <td>3.029476</td>\n",
       "      <td>24.305621</td>\n",
       "      <td>8.686820</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>d29c2c54acc38ff3c0614d0a653813dd</td>\n",
       "      <td>MISSING</td>\n",
       "      <td>4660</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2009-08-21</td>\n",
       "      <td>2016-08-30</td>\n",
       "      <td>2009-08-21</td>\n",
       "      <td>2015-08-31</td>\n",
       "      <td>189.95</td>\n",
       "      <td>...</td>\n",
       "      <td>1.880790</td>\n",
       "      <td>531.736503</td>\n",
       "      <td>-1.770692</td>\n",
       "      <td>-0.000003</td>\n",
       "      <td>-2.947802</td>\n",
       "      <td>-0.000003</td>\n",
       "      <td>1.073045</td>\n",
       "      <td>0.401609</td>\n",
       "      <td>1.073045</td>\n",
       "      <td>0.401609</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>764c75f661154dac3a6c254cd082ea7d</td>\n",
       "      <td>foosdfpfkusacimwkcsosbicdxkicaua</td>\n",
       "      <td>544</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2010-04-16</td>\n",
       "      <td>2016-04-16</td>\n",
       "      <td>2010-04-16</td>\n",
       "      <td>2015-04-17</td>\n",
       "      <td>47.96</td>\n",
       "      <td>...</td>\n",
       "      <td>3.107195</td>\n",
       "      <td>532.625404</td>\n",
       "      <td>-1.354201</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-2.684050</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.401606</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.401606</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>bba03439a292a1e166f80264c16191cb</td>\n",
       "      <td>lmkebamcaaclubfxadlmueccxoimlema</td>\n",
       "      <td>1584</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>2010-03-30</td>\n",
       "      <td>2016-03-30</td>\n",
       "      <td>2010-03-30</td>\n",
       "      <td>2015-03-31</td>\n",
       "      <td>240.04</td>\n",
       "      <td>...</td>\n",
       "      <td>1.814521</td>\n",
       "      <td>532.803183</td>\n",
       "      <td>-1.485466</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-2.915425</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.401606</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.401606</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>149d57cf92fc41cf94415803a877cb4b</td>\n",
       "      <td>MISSING</td>\n",
       "      <td>4425</td>\n",
       "      <td>0</td>\n",
       "      <td>526</td>\n",
       "      <td>2010-01-13</td>\n",
       "      <td>2016-03-07</td>\n",
       "      <td>2010-01-13</td>\n",
       "      <td>2015-03-09</td>\n",
       "      <td>445.75</td>\n",
       "      <td>...</td>\n",
       "      <td>3.608133</td>\n",
       "      <td>976.515746</td>\n",
       "      <td>-6.372446</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-6.372446</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.561484</td>\n",
       "      <td>0.401607</td>\n",
       "      <td>1.561484</td>\n",
       "      <td>0.401607</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>5 rows × 43 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                 id                     channel_sales  \\\n",
       "0  24011ae4ebbe3035111d65fa7c15bc57  foosdfpfkusacimwkcsosbicdxkicaua   \n",
       "1  d29c2c54acc38ff3c0614d0a653813dd                           MISSING   \n",
       "2  764c75f661154dac3a6c254cd082ea7d  foosdfpfkusacimwkcsosbicdxkicaua   \n",
       "3  bba03439a292a1e166f80264c16191cb  lmkebamcaaclubfxadlmueccxoimlema   \n",
       "4  149d57cf92fc41cf94415803a877cb4b                           MISSING   \n",
       "\n",
       "   cons_12m  cons_gas_12m  cons_last_month date_activ   date_end  \\\n",
       "0         0         54946                0 2013-06-15 2016-06-15   \n",
       "1      4660             0                0 2009-08-21 2016-08-30   \n",
       "2       544             0                0 2010-04-16 2016-04-16   \n",
       "3      1584             0                0 2010-03-30 2016-03-30   \n",
       "4      4425             0              526 2010-01-13 2016-03-07   \n",
       "\n",
       "  date_modif_prod date_renewal  forecast_cons_12m  ...  Total_var   Total_fix  \\\n",
       "0      2015-11-01   2015-06-23               0.00  ...   3.504802  938.347394   \n",
       "1      2009-08-21   2015-08-31             189.95  ...   1.880790  531.736503   \n",
       "2      2010-04-16   2015-04-17              47.96  ...   3.107195  532.625404   \n",
       "3      2010-03-30   2015-03-31             240.04  ...   1.814521  532.803183   \n",
       "4      2010-01-13   2015-03-09             445.75  ...   3.608133  976.515746   \n",
       "\n",
       "   off_peak_energy_negative_avg  off_peak_power_negative_avg  \\\n",
       "0                     -3.381130                     0.000000   \n",
       "1                     -1.770692                    -0.000003   \n",
       "2                     -1.354201                     0.000000   \n",
       "3                     -1.485466                     0.000000   \n",
       "4                     -6.372446                     0.000000   \n",
       "\n",
       "   off_peak_energy_negative_max  off_peak_power_negative_max  \\\n",
       "0                     -6.495543                     0.000000   \n",
       "1                     -2.947802                    -0.000003   \n",
       "2                     -2.684050                     0.000000   \n",
       "3                     -2.915425                     0.000000   \n",
       "4                     -6.372446                     0.000000   \n",
       "\n",
       "  off_peak_energy_positive_avg  off_peak_power_positive_avg  \\\n",
       "0                    24.305621                     3.029476   \n",
       "1                     1.073045                     0.401609   \n",
       "2                     0.000000                     0.401606   \n",
       "3                     0.000000                     0.401606   \n",
       "4                     1.561484                     0.401607   \n",
       "\n",
       "   off_peak_energy_positive_max  off_peak_power_positive_max  \n",
       "0                     24.305621                     8.686820  \n",
       "1                      1.073045                     0.401609  \n",
       "2                      0.000000                     0.401606  \n",
       "3                      0.000000                     0.401606  \n",
       "4                      1.561484                     0.401607  \n",
       "\n",
       "[5 rows x 43 columns]"
      ]
     },
     "execution_count": 76,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dft.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "id": "dbd7f15d",
   "metadata": {},
   "outputs": [],
   "source": [
    "#lets store this file which will help us in modelling\n",
    "#dft.to_csv('Feature_Engineered_Data.csv',index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 95,
   "id": "a8424015",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['id', 'channel_sales', 'cons_12m', 'cons_gas_12m', 'cons_last_month',\n",
       "       'date_activ', 'date_end', 'date_modif_prod', 'date_renewal',\n",
       "       'forecast_cons_12m', 'forecast_cons_year', 'forecast_discount_energy',\n",
       "       'forecast_meter_rent_12m', 'forecast_price_energy_off_peak',\n",
       "       'forecast_price_energy_peak', 'forecast_price_pow_off_peak', 'has_gas',\n",
       "       'imp_cons', 'margin_gross_pow_ele', 'margin_net_pow_ele', 'nb_prod_act',\n",
       "       'net_margin', 'num_years_antig', 'origin_up', 'pow_max', 'churn',\n",
       "       'price_off_peak_var', 'price_peak_var', 'price_mid_peak_var',\n",
       "       'price_off_peak_fix', 'price_peak_fix', 'price_mid_peak_fix',\n",
       "       'Month_Count', 'Total_var', 'Total_fix', 'off_peak_energy_negative_avg',\n",
       "       'off_peak_power_negative_avg', 'off_peak_energy_negative_max',\n",
       "       'off_peak_power_negative_max', 'off_peak_energy_positive_avg',\n",
       "       'off_peak_power_positive_avg', 'off_peak_energy_positive_max',\n",
       "       'off_peak_power_positive_max'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 95,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dft.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "396c03dd",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "interpreter": {
   "hash": "152bf6e7dc8ee53edb5af21dc1a8faeab7f134840808a94079ed98d91ece7e0c"
  },
  "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}