dynamic_rfm_II.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import csv\n",
    "import datetime as dt\n",
    "from datetime import date\n",
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "plt.style.use('ggplot')\n",
    "sns.set(font_scale=1.5)\n",
    "%config InlineBackend.figure_format = 'retina'\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.cluster import KMeans\n",
    "from sklearn.preprocessing import StandardScaler"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "file_location = 'C:\\\\Users\\\\Tan\\\\Desktop\\\\GA\\\\Capstone\\\\know_customer_daily_data.csv'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "f = pd.read_csv(file_location)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.DataFrame(f)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['DateTime'] = pd.to_datetime(df['date'], format='%Y-%m-%d')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "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>date</th>\n",
       "      <th>id</th>\n",
       "      <th>invoices_no</th>\n",
       "      <th>categories</th>\n",
       "      <th>quantity</th>\n",
       "      <th>unitprice</th>\n",
       "      <th>gross</th>\n",
       "      <th>top_products</th>\n",
       "      <th>DateTime</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2010-12-01</td>\n",
       "      <td>12748.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>4.950000</td>\n",
       "      <td>4.95</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>2010-12-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2010-12-01</td>\n",
       "      <td>12838.0</td>\n",
       "      <td>1</td>\n",
       "      <td>59</td>\n",
       "      <td>228</td>\n",
       "      <td>1.973390</td>\n",
       "      <td>390.79</td>\n",
       "      <td>0.576271</td>\n",
       "      <td>2010-12-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2010-12-01</td>\n",
       "      <td>12868.0</td>\n",
       "      <td>1</td>\n",
       "      <td>12</td>\n",
       "      <td>112</td>\n",
       "      <td>3.170833</td>\n",
       "      <td>203.30</td>\n",
       "      <td>0.833333</td>\n",
       "      <td>2010-12-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2010-12-01</td>\n",
       "      <td>12921.0</td>\n",
       "      <td>1</td>\n",
       "      <td>15</td>\n",
       "      <td>140</td>\n",
       "      <td>2.483333</td>\n",
       "      <td>322.40</td>\n",
       "      <td>0.933333</td>\n",
       "      <td>2010-12-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2010-12-01</td>\n",
       "      <td>12947.0</td>\n",
       "      <td>1</td>\n",
       "      <td>17</td>\n",
       "      <td>184</td>\n",
       "      <td>2.960588</td>\n",
       "      <td>304.04</td>\n",
       "      <td>0.470588</td>\n",
       "      <td>2010-12-01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         date       id  invoices_no  categories  quantity  unitprice   gross  \\\n",
       "0  2010-12-01  12748.0            1           1         1   4.950000    4.95   \n",
       "1  2010-12-01  12838.0            1          59       228   1.973390  390.79   \n",
       "2  2010-12-01  12868.0            1          12       112   3.170833  203.30   \n",
       "3  2010-12-01  12921.0            1          15       140   2.483333  322.40   \n",
       "4  2010-12-01  12947.0            1          17       184   2.960588  304.04   \n",
       "\n",
       "   top_products   DateTime  \n",
       "0      1.000000 2010-12-01  \n",
       "1      0.576271 2010-12-01  \n",
       "2      0.833333 2010-12-01  \n",
       "3      0.933333 2010-12-01  \n",
       "4      0.470588 2010-12-01  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 17305 entries, 0 to 17304\n",
      "Data columns (total 9 columns):\n",
      "date            17305 non-null object\n",
      "id              17305 non-null float64\n",
      "invoices_no     17305 non-null int64\n",
      "categories      17305 non-null int64\n",
      "quantity        17305 non-null int64\n",
      "unitprice       17305 non-null float64\n",
      "gross           17305 non-null float64\n",
      "top_products    17305 non-null float64\n",
      "DateTime        17305 non-null datetime64[ns]\n",
      "dtypes: datetime64[ns](1), float64(4), int64(3), object(1)\n",
      "memory usage: 1.2+ MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Apply RFM customer segmentation model on monthly periodicity, using a 90d rolling window.\n",
    "# Recency would be number of days since last transaction\n",
    "# Monetary would be total gross revenues over 90 days\n",
    "# Frequency would be the number of transactions over 90 days\n",
    "# Finally, aggregate the daily data to montly data..."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Create \"days_since_last\" variable"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# \"days_since_last\" indicates how many days since last transaction\n",
    "df['days_since_last'] = (df.groupby('id')['DateTime']\n",
    "                 .apply(lambda x: (x - x.shift(1)).dt.days))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>date</th>\n",
       "      <th>id</th>\n",
       "      <th>invoices_no</th>\n",
       "      <th>categories</th>\n",
       "      <th>quantity</th>\n",
       "      <th>unitprice</th>\n",
       "      <th>gross</th>\n",
       "      <th>top_products</th>\n",
       "      <th>DateTime</th>\n",
       "      <th>days_since_last</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1426</th>\n",
       "      <td>2011-01-06</td>\n",
       "      <td>18283.0</td>\n",
       "      <td>1</td>\n",
       "      <td>57</td>\n",
       "      <td>61</td>\n",
       "      <td>1.771053</td>\n",
       "      <td>108.45</td>\n",
       "      <td>0.771930</td>\n",
       "      <td>2011-01-06</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1941</th>\n",
       "      <td>2011-01-23</td>\n",
       "      <td>18283.0</td>\n",
       "      <td>1</td>\n",
       "      <td>45</td>\n",
       "      <td>67</td>\n",
       "      <td>1.850000</td>\n",
       "      <td>106.55</td>\n",
       "      <td>0.644444</td>\n",
       "      <td>2011-01-23</td>\n",
       "      <td>17.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3222</th>\n",
       "      <td>2011-02-28</td>\n",
       "      <td>18283.0</td>\n",
       "      <td>1</td>\n",
       "      <td>30</td>\n",
       "      <td>59</td>\n",
       "      <td>2.351667</td>\n",
       "      <td>102.90</td>\n",
       "      <td>0.533333</td>\n",
       "      <td>2011-02-28</td>\n",
       "      <td>36.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5450</th>\n",
       "      <td>2011-04-21</td>\n",
       "      <td>18283.0</td>\n",
       "      <td>1</td>\n",
       "      <td>56</td>\n",
       "      <td>87</td>\n",
       "      <td>1.478929</td>\n",
       "      <td>117.68</td>\n",
       "      <td>0.714286</td>\n",
       "      <td>2011-04-21</td>\n",
       "      <td>52.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6732</th>\n",
       "      <td>2011-05-23</td>\n",
       "      <td>18283.0</td>\n",
       "      <td>1</td>\n",
       "      <td>44</td>\n",
       "      <td>62</td>\n",
       "      <td>1.710000</td>\n",
       "      <td>99.47</td>\n",
       "      <td>0.909091</td>\n",
       "      <td>2011-05-23</td>\n",
       "      <td>32.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7685</th>\n",
       "      <td>2011-06-14</td>\n",
       "      <td>18283.0</td>\n",
       "      <td>1</td>\n",
       "      <td>24</td>\n",
       "      <td>52</td>\n",
       "      <td>2.042917</td>\n",
       "      <td>103.72</td>\n",
       "      <td>0.875000</td>\n",
       "      <td>2011-06-14</td>\n",
       "      <td>22.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8115</th>\n",
       "      <td>2011-06-23</td>\n",
       "      <td>18283.0</td>\n",
       "      <td>1</td>\n",
       "      <td>87</td>\n",
       "      <td>157</td>\n",
       "      <td>1.526667</td>\n",
       "      <td>203.81</td>\n",
       "      <td>0.793103</td>\n",
       "      <td>2011-06-23</td>\n",
       "      <td>9.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8908</th>\n",
       "      <td>2011-07-14</td>\n",
       "      <td>18283.0</td>\n",
       "      <td>2</td>\n",
       "      <td>57</td>\n",
       "      <td>102</td>\n",
       "      <td>1.627368</td>\n",
       "      <td>143.19</td>\n",
       "      <td>0.807018</td>\n",
       "      <td>2011-07-14</td>\n",
       "      <td>21.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11051</th>\n",
       "      <td>2011-09-05</td>\n",
       "      <td>18283.0</td>\n",
       "      <td>1</td>\n",
       "      <td>47</td>\n",
       "      <td>95</td>\n",
       "      <td>1.401915</td>\n",
       "      <td>134.90</td>\n",
       "      <td>0.851064</td>\n",
       "      <td>2011-09-05</td>\n",
       "      <td>53.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13992</th>\n",
       "      <td>2011-10-27</td>\n",
       "      <td>18283.0</td>\n",
       "      <td>1</td>\n",
       "      <td>40</td>\n",
       "      <td>63</td>\n",
       "      <td>2.087000</td>\n",
       "      <td>114.65</td>\n",
       "      <td>0.675000</td>\n",
       "      <td>2011-10-27</td>\n",
       "      <td>52.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15021</th>\n",
       "      <td>2011-11-10</td>\n",
       "      <td>18283.0</td>\n",
       "      <td>2</td>\n",
       "      <td>60</td>\n",
       "      <td>65</td>\n",
       "      <td>1.760833</td>\n",
       "      <td>114.30</td>\n",
       "      <td>0.883333</td>\n",
       "      <td>2011-11-10</td>\n",
       "      <td>14.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16067</th>\n",
       "      <td>2011-11-23</td>\n",
       "      <td>18283.0</td>\n",
       "      <td>1</td>\n",
       "      <td>107</td>\n",
       "      <td>251</td>\n",
       "      <td>1.196542</td>\n",
       "      <td>313.65</td>\n",
       "      <td>0.654206</td>\n",
       "      <td>2011-11-23</td>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16577</th>\n",
       "      <td>2011-11-30</td>\n",
       "      <td>18283.0</td>\n",
       "      <td>1</td>\n",
       "      <td>52</td>\n",
       "      <td>134</td>\n",
       "      <td>1.636154</td>\n",
       "      <td>223.61</td>\n",
       "      <td>0.923077</td>\n",
       "      <td>2011-11-30</td>\n",
       "      <td>7.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17075</th>\n",
       "      <td>2011-12-06</td>\n",
       "      <td>18283.0</td>\n",
       "      <td>1</td>\n",
       "      <td>50</td>\n",
       "      <td>142</td>\n",
       "      <td>1.307600</td>\n",
       "      <td>208.00</td>\n",
       "      <td>0.680000</td>\n",
       "      <td>2011-12-06</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             date       id  invoices_no  categories  quantity  unitprice  \\\n",
       "1426   2011-01-06  18283.0            1          57        61   1.771053   \n",
       "1941   2011-01-23  18283.0            1          45        67   1.850000   \n",
       "3222   2011-02-28  18283.0            1          30        59   2.351667   \n",
       "5450   2011-04-21  18283.0            1          56        87   1.478929   \n",
       "6732   2011-05-23  18283.0            1          44        62   1.710000   \n",
       "7685   2011-06-14  18283.0            1          24        52   2.042917   \n",
       "8115   2011-06-23  18283.0            1          87       157   1.526667   \n",
       "8908   2011-07-14  18283.0            2          57       102   1.627368   \n",
       "11051  2011-09-05  18283.0            1          47        95   1.401915   \n",
       "13992  2011-10-27  18283.0            1          40        63   2.087000   \n",
       "15021  2011-11-10  18283.0            2          60        65   1.760833   \n",
       "16067  2011-11-23  18283.0            1         107       251   1.196542   \n",
       "16577  2011-11-30  18283.0            1          52       134   1.636154   \n",
       "17075  2011-12-06  18283.0            1          50       142   1.307600   \n",
       "\n",
       "        gross  top_products   DateTime  days_since_last  \n",
       "1426   108.45      0.771930 2011-01-06              NaN  \n",
       "1941   106.55      0.644444 2011-01-23             17.0  \n",
       "3222   102.90      0.533333 2011-02-28             36.0  \n",
       "5450   117.68      0.714286 2011-04-21             52.0  \n",
       "6732    99.47      0.909091 2011-05-23             32.0  \n",
       "7685   103.72      0.875000 2011-06-14             22.0  \n",
       "8115   203.81      0.793103 2011-06-23              9.0  \n",
       "8908   143.19      0.807018 2011-07-14             21.0  \n",
       "11051  134.90      0.851064 2011-09-05             53.0  \n",
       "13992  114.65      0.675000 2011-10-27             52.0  \n",
       "15021  114.30      0.883333 2011-11-10             14.0  \n",
       "16067  313.65      0.654206 2011-11-23             13.0  \n",
       "16577  223.61      0.923077 2011-11-30              7.0  \n",
       "17075  208.00      0.680000 2011-12-06              6.0  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check if new variable is good, using one customer example\n",
    "df[df['id']==18283]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Create the \"frequency\" & \"monetary variables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.set_index('DateTime', inplace=True, drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create dates for entire sample period, which needs to be in same format as the \"date\" in df\n",
    "dates = [i.date() for i in pd.date_range(start='12/01/2010', end='12/09/2011')] "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "temp_freq = (df.groupby('id')['invoices_no']\n",
    "    .apply(lambda x: x.reindex(dates) \n",
    "                      .fillna(0) \n",
    "                      .shift() # Not including the current day\n",
    "                      .rolling(90)\n",
    "                      .sum()) \n",
    "    .reset_index()\n",
    "    .rename(columns={'invoices_no': 'trans_90d', \n",
    "                     'DateTime': 'date'}))\n",
    "\n",
    "temp_freq['date'] = pd.to_datetime(temp_freq['date'], format='%Y-%m-%d', utc=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "temp_money = (df.groupby('id')['gross']\n",
    "    .apply(lambda x: x.reindex(dates)  \n",
    "                      .fillna(0)\n",
    "                      .shift() \n",
    "                      .rolling(90)\n",
    "                      .sum()) \n",
    "    .reset_index()\n",
    "    .rename(columns={'gross': 'revenue_90d', \n",
    "                     'DateTime': 'date'}))\n",
    "\n",
    "temp_money['date'] = pd.to_datetime(temp_money['date'], format='%Y-%m-%d', utc=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Merging on datetime values can be tricky. Need to set utc=True beforehand.\n",
    "df['date'] = pd.to_datetime(df['date'], format='%Y-%m-%d', utc=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.merge(temp_freq, on=['id', 'date'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.merge(temp_money, on=['id', 'date'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Average gross revenues might be a good alternative Monetary variable, rather than total gross revenues...\n",
    "# Generating an average gross revenue variable:\n",
    "def avg_gross(row):\n",
    "    if row['trans_90d'] == 0:\n",
    "        return 0\n",
    "    elif row['revenue_90d'] == 0:\n",
    "        return 0\n",
    "    else:\n",
    "        return row['revenue_90d']/row['trans_90d']\n",
    "    \n",
    "df['avg_revenue_90d'] = df.apply(lambda row: avg_gross(row), axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "Int64Index: 17305 entries, 0 to 17304\n",
      "Data columns (total 12 columns):\n",
      "date               17305 non-null datetime64[ns, UTC]\n",
      "id                 17305 non-null float64\n",
      "invoices_no        17305 non-null int64\n",
      "categories         17305 non-null int64\n",
      "quantity           17305 non-null int64\n",
      "unitprice          17305 non-null float64\n",
      "gross              17305 non-null float64\n",
      "top_products       17305 non-null float64\n",
      "days_since_last    13358 non-null float64\n",
      "trans_90d          14082 non-null float64\n",
      "revenue_90d        14082 non-null float64\n",
      "avg_revenue_90d    14082 non-null float64\n",
      "dtypes: datetime64[ns, UTC](1), float64(8), int64(3)\n",
      "memory usage: 1.7 MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "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>date</th>\n",
       "      <th>id</th>\n",
       "      <th>invoices_no</th>\n",
       "      <th>categories</th>\n",
       "      <th>quantity</th>\n",
       "      <th>unitprice</th>\n",
       "      <th>gross</th>\n",
       "      <th>top_products</th>\n",
       "      <th>days_since_last</th>\n",
       "      <th>trans_90d</th>\n",
       "      <th>revenue_90d</th>\n",
       "      <th>avg_revenue_90d</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>2010-12-01 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>27.500000</td>\n",
       "      <td>-27.50</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>263</th>\n",
       "      <td>2010-12-05 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>2</td>\n",
       "      <td>29</td>\n",
       "      <td>59</td>\n",
       "      <td>6.180690</td>\n",
       "      <td>263.66</td>\n",
       "      <td>0.655172</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>602</th>\n",
       "      <td>2010-12-09 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>1</td>\n",
       "      <td>29</td>\n",
       "      <td>48</td>\n",
       "      <td>4.891034</td>\n",
       "      <td>224.38</td>\n",
       "      <td>0.689655</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>952</th>\n",
       "      <td>2010-12-15 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>2</td>\n",
       "      <td>30</td>\n",
       "      <td>50</td>\n",
       "      <td>6.169333</td>\n",
       "      <td>242.38</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>6.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1194</th>\n",
       "      <td>2010-12-20 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>2</td>\n",
       "      <td>20</td>\n",
       "      <td>23</td>\n",
       "      <td>6.624000</td>\n",
       "      <td>124.97</td>\n",
       "      <td>0.800000</td>\n",
       "      <td>5.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1337</th>\n",
       "      <td>2011-01-05 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>2</td>\n",
       "      <td>25</td>\n",
       "      <td>32</td>\n",
       "      <td>7.820800</td>\n",
       "      <td>206.78</td>\n",
       "      <td>0.520000</td>\n",
       "      <td>16.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1520</th>\n",
       "      <td>2011-01-10 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>1</td>\n",
       "      <td>21</td>\n",
       "      <td>37</td>\n",
       "      <td>5.976190</td>\n",
       "      <td>167.60</td>\n",
       "      <td>0.714286</td>\n",
       "      <td>5.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1820</th>\n",
       "      <td>2011-01-19 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>1</td>\n",
       "      <td>19</td>\n",
       "      <td>45</td>\n",
       "      <td>6.050000</td>\n",
       "      <td>172.15</td>\n",
       "      <td>0.842105</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2005</th>\n",
       "      <td>2011-01-25 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>14</td>\n",
       "      <td>7.354545</td>\n",
       "      <td>90.90</td>\n",
       "      <td>0.909091</td>\n",
       "      <td>6.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2222</th>\n",
       "      <td>2011-01-31 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>1</td>\n",
       "      <td>14</td>\n",
       "      <td>22</td>\n",
       "      <td>6.232143</td>\n",
       "      <td>101.80</td>\n",
       "      <td>0.428571</td>\n",
       "      <td>6.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2499</th>\n",
       "      <td>2011-02-08 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>1</td>\n",
       "      <td>15</td>\n",
       "      <td>24</td>\n",
       "      <td>6.616667</td>\n",
       "      <td>140.55</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>8.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2700</th>\n",
       "      <td>2011-02-14 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>70.010000</td>\n",
       "      <td>-70.01</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>6.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2736</th>\n",
       "      <td>2011-02-15 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>2</td>\n",
       "      <td>16</td>\n",
       "      <td>30</td>\n",
       "      <td>3.931250</td>\n",
       "      <td>103.95</td>\n",
       "      <td>0.562500</td>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3078</th>\n",
       "      <td>2011-02-24 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>16</td>\n",
       "      <td>5.544444</td>\n",
       "      <td>77.95</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>9.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3195</th>\n",
       "      <td>2011-02-28 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>1</td>\n",
       "      <td>11</td>\n",
       "      <td>20</td>\n",
       "      <td>7.040909</td>\n",
       "      <td>122.00</td>\n",
       "      <td>0.727273</td>\n",
       "      <td>4.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3337</th>\n",
       "      <td>2011-03-03 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>30.380000</td>\n",
       "      <td>-30.38</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>1969.06</td>\n",
       "      <td>103.634737</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3557</th>\n",
       "      <td>2011-03-09 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>2</td>\n",
       "      <td>12</td>\n",
       "      <td>14</td>\n",
       "      <td>7.351667</td>\n",
       "      <td>88.83</td>\n",
       "      <td>0.583333</td>\n",
       "      <td>6.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1675.02</td>\n",
       "      <td>93.056667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3764</th>\n",
       "      <td>2011-03-15 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>1</td>\n",
       "      <td>12</td>\n",
       "      <td>23</td>\n",
       "      <td>3.831667</td>\n",
       "      <td>78.04</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>6.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>1539.47</td>\n",
       "      <td>81.024737</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4165</th>\n",
       "      <td>2011-03-24 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>1</td>\n",
       "      <td>21</td>\n",
       "      <td>27</td>\n",
       "      <td>2.640000</td>\n",
       "      <td>73.77</td>\n",
       "      <td>0.476190</td>\n",
       "      <td>9.0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>1250.16</td>\n",
       "      <td>78.135000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4470</th>\n",
       "      <td>2011-03-31 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>2</td>\n",
       "      <td>24</td>\n",
       "      <td>54</td>\n",
       "      <td>3.786667</td>\n",
       "      <td>127.99</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>7.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>1323.93</td>\n",
       "      <td>77.878235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4746</th>\n",
       "      <td>2011-04-07 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>2</td>\n",
       "      <td>8</td>\n",
       "      <td>13</td>\n",
       "      <td>4.161250</td>\n",
       "      <td>28.81</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>7.0</td>\n",
       "      <td>17.0</td>\n",
       "      <td>1245.14</td>\n",
       "      <td>73.243529</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5152</th>\n",
       "      <td>2011-04-15 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>11</td>\n",
       "      <td>5.607143</td>\n",
       "      <td>58.85</td>\n",
       "      <td>0.571429</td>\n",
       "      <td>8.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1106.35</td>\n",
       "      <td>61.463889</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5341</th>\n",
       "      <td>2011-04-20 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>1</td>\n",
       "      <td>12</td>\n",
       "      <td>25</td>\n",
       "      <td>5.135000</td>\n",
       "      <td>103.15</td>\n",
       "      <td>0.583333</td>\n",
       "      <td>5.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>993.05</td>\n",
       "      <td>55.169444</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5526</th>\n",
       "      <td>2011-04-27 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>2</td>\n",
       "      <td>11</td>\n",
       "      <td>54</td>\n",
       "      <td>6.156364</td>\n",
       "      <td>105.90</td>\n",
       "      <td>0.363636</td>\n",
       "      <td>7.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1005.30</td>\n",
       "      <td>55.850000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5972</th>\n",
       "      <td>2011-05-09 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>2</td>\n",
       "      <td>19</td>\n",
       "      <td>30</td>\n",
       "      <td>4.704211</td>\n",
       "      <td>106.65</td>\n",
       "      <td>0.789474</td>\n",
       "      <td>12.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>1009.40</td>\n",
       "      <td>53.126316</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6225</th>\n",
       "      <td>2011-05-13 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>1</td>\n",
       "      <td>14</td>\n",
       "      <td>23</td>\n",
       "      <td>4.219286</td>\n",
       "      <td>90.84</td>\n",
       "      <td>0.857143</td>\n",
       "      <td>4.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>975.50</td>\n",
       "      <td>48.775000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6358</th>\n",
       "      <td>2011-05-17 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>-1</td>\n",
       "      <td>9.840000</td>\n",
       "      <td>-9.84</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1032.40</td>\n",
       "      <td>57.355556</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6431</th>\n",
       "      <td>2011-05-18 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>2</td>\n",
       "      <td>18</td>\n",
       "      <td>95</td>\n",
       "      <td>3.402778</td>\n",
       "      <td>143.69</td>\n",
       "      <td>0.722222</td>\n",
       "      <td>1.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>1022.56</td>\n",
       "      <td>53.818947</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6757</th>\n",
       "      <td>2011-05-24 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>2</td>\n",
       "      <td>15</td>\n",
       "      <td>34</td>\n",
       "      <td>6.649333</td>\n",
       "      <td>124.99</td>\n",
       "      <td>0.733333</td>\n",
       "      <td>6.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>1166.25</td>\n",
       "      <td>55.535714</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7378</th>\n",
       "      <td>2011-06-08 00:00:00+00:00</td>\n",
       "      <td>14527.0</td>\n",
       "      <td>2</td>\n",
       "      <td>26</td>\n",
       "      <td>48</td>\n",
       "      <td>6.407308</td>\n",
       "      <td>188.36</td>\n",
       "      <td>0.730769</td>\n",
       "      <td>15.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1032.84</td>\n",
       "      <td>57.380000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                          date       id  invoices_no  categories  quantity  \\\n",
       "27   2010-12-01 00:00:00+00:00  14527.0            1           1        -1   \n",
       "263  2010-12-05 00:00:00+00:00  14527.0            2          29        59   \n",
       "602  2010-12-09 00:00:00+00:00  14527.0            1          29        48   \n",
       "952  2010-12-15 00:00:00+00:00  14527.0            2          30        50   \n",
       "1194 2010-12-20 00:00:00+00:00  14527.0            2          20        23   \n",
       "1337 2011-01-05 00:00:00+00:00  14527.0            2          25        32   \n",
       "1520 2011-01-10 00:00:00+00:00  14527.0            1          21        37   \n",
       "1820 2011-01-19 00:00:00+00:00  14527.0            1          19        45   \n",
       "2005 2011-01-25 00:00:00+00:00  14527.0            1          11        14   \n",
       "2222 2011-01-31 00:00:00+00:00  14527.0            1          14        22   \n",
       "2499 2011-02-08 00:00:00+00:00  14527.0            1          15        24   \n",
       "2700 2011-02-14 00:00:00+00:00  14527.0            1           1        -1   \n",
       "2736 2011-02-15 00:00:00+00:00  14527.0            2          16        30   \n",
       "3078 2011-02-24 00:00:00+00:00  14527.0            1           9        16   \n",
       "3195 2011-02-28 00:00:00+00:00  14527.0            1          11        20   \n",
       "3337 2011-03-03 00:00:00+00:00  14527.0            1           1        -1   \n",
       "3557 2011-03-09 00:00:00+00:00  14527.0            2          12        14   \n",
       "3764 2011-03-15 00:00:00+00:00  14527.0            1          12        23   \n",
       "4165 2011-03-24 00:00:00+00:00  14527.0            1          21        27   \n",
       "4470 2011-03-31 00:00:00+00:00  14527.0            2          24        54   \n",
       "4746 2011-04-07 00:00:00+00:00  14527.0            2           8        13   \n",
       "5152 2011-04-15 00:00:00+00:00  14527.0            1           7        11   \n",
       "5341 2011-04-20 00:00:00+00:00  14527.0            1          12        25   \n",
       "5526 2011-04-27 00:00:00+00:00  14527.0            2          11        54   \n",
       "5972 2011-05-09 00:00:00+00:00  14527.0            2          19        30   \n",
       "6225 2011-05-13 00:00:00+00:00  14527.0            1          14        23   \n",
       "6358 2011-05-17 00:00:00+00:00  14527.0            1           1        -1   \n",
       "6431 2011-05-18 00:00:00+00:00  14527.0            2          18        95   \n",
       "6757 2011-05-24 00:00:00+00:00  14527.0            2          15        34   \n",
       "7378 2011-06-08 00:00:00+00:00  14527.0            2          26        48   \n",
       "\n",
       "      unitprice   gross  top_products  days_since_last  trans_90d  \\\n",
       "27    27.500000  -27.50      0.000000              NaN        NaN   \n",
       "263    6.180690  263.66      0.655172              4.0        NaN   \n",
       "602    4.891034  224.38      0.689655              4.0        NaN   \n",
       "952    6.169333  242.38      0.666667              6.0        NaN   \n",
       "1194   6.624000  124.97      0.800000              5.0        NaN   \n",
       "1337   7.820800  206.78      0.520000             16.0        NaN   \n",
       "1520   5.976190  167.60      0.714286              5.0        NaN   \n",
       "1820   6.050000  172.15      0.842105              9.0        NaN   \n",
       "2005   7.354545   90.90      0.909091              6.0        NaN   \n",
       "2222   6.232143  101.80      0.428571              6.0        NaN   \n",
       "2499   6.616667  140.55      0.666667              8.0        NaN   \n",
       "2700  70.010000  -70.01      0.000000              6.0        NaN   \n",
       "2736   3.931250  103.95      0.562500              1.0        NaN   \n",
       "3078   5.544444   77.95      0.666667              9.0        NaN   \n",
       "3195   7.040909  122.00      0.727273              4.0        NaN   \n",
       "3337  30.380000  -30.38      0.000000              3.0       19.0   \n",
       "3557   7.351667   88.83      0.583333              6.0       18.0   \n",
       "3764   3.831667   78.04      0.750000              6.0       19.0   \n",
       "4165   2.640000   73.77      0.476190              9.0       16.0   \n",
       "4470   3.786667  127.99      0.333333              7.0       17.0   \n",
       "4746   4.161250   28.81      0.500000              7.0       17.0   \n",
       "5152   5.607143   58.85      0.571429              8.0       18.0   \n",
       "5341   5.135000  103.15      0.583333              5.0       18.0   \n",
       "5526   6.156364  105.90      0.363636              7.0       18.0   \n",
       "5972   4.704211  106.65      0.789474             12.0       19.0   \n",
       "6225   4.219286   90.84      0.857143              4.0       20.0   \n",
       "6358   9.840000   -9.84      0.000000              4.0       18.0   \n",
       "6431   3.402778  143.69      0.722222              1.0       19.0   \n",
       "6757   6.649333  124.99      0.733333              6.0       21.0   \n",
       "7378   6.407308  188.36      0.730769             15.0       18.0   \n",
       "\n",
       "      revenue_90d  avg_revenue_90d  \n",
       "27            NaN              NaN  \n",
       "263           NaN              NaN  \n",
       "602           NaN              NaN  \n",
       "952           NaN              NaN  \n",
       "1194          NaN              NaN  \n",
       "1337          NaN              NaN  \n",
       "1520          NaN              NaN  \n",
       "1820          NaN              NaN  \n",
       "2005          NaN              NaN  \n",
       "2222          NaN              NaN  \n",
       "2499          NaN              NaN  \n",
       "2700          NaN              NaN  \n",
       "2736          NaN              NaN  \n",
       "3078          NaN              NaN  \n",
       "3195          NaN              NaN  \n",
       "3337      1969.06       103.634737  \n",
       "3557      1675.02        93.056667  \n",
       "3764      1539.47        81.024737  \n",
       "4165      1250.16        78.135000  \n",
       "4470      1323.93        77.878235  \n",
       "4746      1245.14        73.243529  \n",
       "5152      1106.35        61.463889  \n",
       "5341       993.05        55.169444  \n",
       "5526      1005.30        55.850000  \n",
       "5972      1009.40        53.126316  \n",
       "6225       975.50        48.775000  \n",
       "6358      1032.40        57.355556  \n",
       "6431      1022.56        53.818947  \n",
       "6757      1166.25        55.535714  \n",
       "7378      1032.84        57.380000  "
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Check if \"frequency\" & \"monetary\" variables are good on a random customer...\n",
    "df[df.id==14527].head(30)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Now to drop all the null values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df[~df['days_since_last'].isnull()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df[~df['revenue_90d'].isnull()]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "False"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# No null values in DF\n",
    "df.isnull().values.any()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Aggregating to monthly data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [],
   "source": [
    "# These three steps create a DateTime index, then transforms the index to only month-end dates\n",
    "df['dateindex'] = pd.to_datetime(df['date'], format='%Y-%m-%d')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.set_index('dateindex', inplace=True, drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Tan\\Anaconda3\\lib\\site-packages\\pandas\\core\\arrays\\datetimes.py:1269: UserWarning: Converting to PeriodArray/Index representation will drop timezone information.\n",
      "  UserWarning,\n"
     ]
    }
   ],
   "source": [
    "df.index = df.index.to_period('M').to_timestamp('M')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "DatetimeIndex: 11706 entries, 2011-03-31 to 2011-12-31\n",
      "Data columns (total 12 columns):\n",
      "date               11706 non-null datetime64[ns, UTC]\n",
      "id                 11706 non-null float64\n",
      "invoices_no        11706 non-null int64\n",
      "categories         11706 non-null int64\n",
      "quantity           11706 non-null int64\n",
      "unitprice          11706 non-null float64\n",
      "gross              11706 non-null float64\n",
      "top_products       11706 non-null float64\n",
      "days_since_last    11706 non-null float64\n",
      "trans_90d          11706 non-null float64\n",
      "revenue_90d        11706 non-null float64\n",
      "avg_revenue_90d    11706 non-null float64\n",
      "dtypes: datetime64[ns, UTC](1), float64(8), int64(3)\n",
      "memory usage: 1.2 MB\n"
     ]
    }
   ],
   "source": [
    "df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['trans_date'] = df['date'].apply(lambda x:x.date())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "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>date</th>\n",
       "      <th>id</th>\n",
       "      <th>invoices_no</th>\n",
       "      <th>categories</th>\n",
       "      <th>quantity</th>\n",
       "      <th>unitprice</th>\n",
       "      <th>gross</th>\n",
       "      <th>top_products</th>\n",
       "      <th>days_since_last</th>\n",
       "      <th>trans_90d</th>\n",
       "      <th>revenue_90d</th>\n",
       "      <th>avg_revenue_90d</th>\n",
       "      <th>trans_date</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dateindex</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2011-03-31</th>\n",
       "      <td>2011-03-01 00:00:00+00:00</td>\n",
       "      <td>12747.0</td>\n",
       "      <td>1</td>\n",
       "      <td>12</td>\n",
       "      <td>146</td>\n",
       "      <td>3.106667</td>\n",
       "      <td>310.78</td>\n",
       "      <td>0.916667</td>\n",
       "      <td>40.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1009.31</td>\n",
       "      <td>336.436667</td>\n",
       "      <td>2011-03-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-03-31</th>\n",
       "      <td>2011-03-01 00:00:00+00:00</td>\n",
       "      <td>12971.0</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>136</td>\n",
       "      <td>1.620000</td>\n",
       "      <td>161.80</td>\n",
       "      <td>0.666667</td>\n",
       "      <td>5.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>1636.07</td>\n",
       "      <td>86.108947</td>\n",
       "      <td>2011-03-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-03-31</th>\n",
       "      <td>2011-03-01 00:00:00+00:00</td>\n",
       "      <td>13110.0</td>\n",
       "      <td>1</td>\n",
       "      <td>27</td>\n",
       "      <td>68</td>\n",
       "      <td>5.269630</td>\n",
       "      <td>268.60</td>\n",
       "      <td>0.740741</td>\n",
       "      <td>19.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>235.75</td>\n",
       "      <td>117.875000</td>\n",
       "      <td>2011-03-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-03-31</th>\n",
       "      <td>2011-03-01 00:00:00+00:00</td>\n",
       "      <td>13599.0</td>\n",
       "      <td>1</td>\n",
       "      <td>15</td>\n",
       "      <td>349</td>\n",
       "      <td>2.695333</td>\n",
       "      <td>580.57</td>\n",
       "      <td>0.733333</td>\n",
       "      <td>19.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>683.30</td>\n",
       "      <td>97.614286</td>\n",
       "      <td>2011-03-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-03-31</th>\n",
       "      <td>2011-03-01 00:00:00+00:00</td>\n",
       "      <td>13880.0</td>\n",
       "      <td>1</td>\n",
       "      <td>9</td>\n",
       "      <td>94</td>\n",
       "      <td>2.616667</td>\n",
       "      <td>200.10</td>\n",
       "      <td>0.777778</td>\n",
       "      <td>13.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>308.00</td>\n",
       "      <td>154.000000</td>\n",
       "      <td>2011-03-01</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                date       id  invoices_no  categories  \\\n",
       "dateindex                                                                \n",
       "2011-03-31 2011-03-01 00:00:00+00:00  12747.0            1          12   \n",
       "2011-03-31 2011-03-01 00:00:00+00:00  12971.0            1           3   \n",
       "2011-03-31 2011-03-01 00:00:00+00:00  13110.0            1          27   \n",
       "2011-03-31 2011-03-01 00:00:00+00:00  13599.0            1          15   \n",
       "2011-03-31 2011-03-01 00:00:00+00:00  13880.0            1           9   \n",
       "\n",
       "            quantity  unitprice   gross  top_products  days_since_last  \\\n",
       "dateindex                                                                \n",
       "2011-03-31       146   3.106667  310.78      0.916667             40.0   \n",
       "2011-03-31       136   1.620000  161.80      0.666667              5.0   \n",
       "2011-03-31        68   5.269630  268.60      0.740741             19.0   \n",
       "2011-03-31       349   2.695333  580.57      0.733333             19.0   \n",
       "2011-03-31        94   2.616667  200.10      0.777778             13.0   \n",
       "\n",
       "            trans_90d  revenue_90d  avg_revenue_90d  trans_date  \n",
       "dateindex                                                        \n",
       "2011-03-31        3.0      1009.31       336.436667  2011-03-01  \n",
       "2011-03-31       19.0      1636.07        86.108947  2011-03-01  \n",
       "2011-03-31        2.0       235.75       117.875000  2011-03-01  \n",
       "2011-03-31        7.0       683.30        97.614286  2011-03-01  \n",
       "2011-03-31        2.0       308.00       154.000000  2011-03-01  "
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['cust_id'] = df['id']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "id_grouping = df.groupby([df.index.year, df.index.month, 'cust_id']).agg({'id': 'last', 'invoices_no': 'sum', \n",
    "                                        'categories': 'mean', 'quantity': 'sum', 'unitprice': 'mean', 'gross': 'sum', \n",
    "                                        'top_products': 'mean', 'days_since_last': 'last',  'trans_90d': 'last', \n",
    "                                        'revenue_90d': 'last', 'avg_revenue_90d': 'last', 'trans_date': 'last'})"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [],
   "source": [
    "# The variables record the transactions of each individual customer per month\n",
    "# sum \"invoices_no\" will give the total number of transactions per month\n",
    "# mean \"categories\" will give the average number of product categories purchased per transaction per month\n",
    "# sum \"quantity\" will give the total number of individual items purchased per month\n",
    "# mean \"unitprice\" will give the average price of the individual items purchased per month\n",
    "# sum \"gross\" will give the total gross spending per month\n",
    "# mean \"top_products\" will give the proportion of top 20% products in the monthly purchases"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [],
   "source": [
    "id_grouping['dateindex'] = pd.to_datetime(id_grouping['trans_date'], format='%Y-%m-%d')\n",
    "id_grouping.set_index('dateindex', inplace=True, drop=True)\n",
    "id_grouping.index = id_grouping.index.to_period('M').to_timestamp('M')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "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>invoices_no</th>\n",
       "      <th>categories</th>\n",
       "      <th>quantity</th>\n",
       "      <th>unitprice</th>\n",
       "      <th>gross</th>\n",
       "      <th>top_products</th>\n",
       "      <th>days_since_last</th>\n",
       "      <th>trans_90d</th>\n",
       "      <th>revenue_90d</th>\n",
       "      <th>avg_revenue_90d</th>\n",
       "      <th>trans_date</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dateindex</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2011-03-31</th>\n",
       "      <td>12747.0</td>\n",
       "      <td>1</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>146</td>\n",
       "      <td>3.106667</td>\n",
       "      <td>310.78</td>\n",
       "      <td>0.916667</td>\n",
       "      <td>40.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1009.31</td>\n",
       "      <td>336.436667</td>\n",
       "      <td>2011-03-01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-03-31</th>\n",
       "      <td>12748.0</td>\n",
       "      <td>13</td>\n",
       "      <td>35.166667</td>\n",
       "      <td>807</td>\n",
       "      <td>1.928548</td>\n",
       "      <td>1011.94</td>\n",
       "      <td>0.428384</td>\n",
       "      <td>1.0</td>\n",
       "      <td>18.0</td>\n",
       "      <td>1439.61</td>\n",
       "      <td>79.978333</td>\n",
       "      <td>2011-03-25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-03-31</th>\n",
       "      <td>12823.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>60</td>\n",
       "      <td>7.650000</td>\n",
       "      <td>459.00</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>35.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>535.50</td>\n",
       "      <td>267.750000</td>\n",
       "      <td>2011-03-30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-03-31</th>\n",
       "      <td>12839.0</td>\n",
       "      <td>1</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-2</td>\n",
       "      <td>1.650000</td>\n",
       "      <td>-3.30</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>10.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1089.01</td>\n",
       "      <td>363.003333</td>\n",
       "      <td>2011-03-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-03-31</th>\n",
       "      <td>12841.0</td>\n",
       "      <td>2</td>\n",
       "      <td>14.500000</td>\n",
       "      <td>191</td>\n",
       "      <td>1.516875</td>\n",
       "      <td>269.90</td>\n",
       "      <td>0.401786</td>\n",
       "      <td>20.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>986.33</td>\n",
       "      <td>140.904286</td>\n",
       "      <td>2011-03-27</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 id  invoices_no  categories  quantity  unitprice    gross  \\\n",
       "dateindex                                                                    \n",
       "2011-03-31  12747.0            1   12.000000       146   3.106667   310.78   \n",
       "2011-03-31  12748.0           13   35.166667       807   1.928548  1011.94   \n",
       "2011-03-31  12823.0            1    1.000000        60   7.650000   459.00   \n",
       "2011-03-31  12839.0            1    1.000000        -2   1.650000    -3.30   \n",
       "2011-03-31  12841.0            2   14.500000       191   1.516875   269.90   \n",
       "\n",
       "            top_products  days_since_last  trans_90d  revenue_90d  \\\n",
       "dateindex                                                           \n",
       "2011-03-31      0.916667             40.0        3.0      1009.31   \n",
       "2011-03-31      0.428384              1.0       18.0      1439.61   \n",
       "2011-03-31      1.000000             35.0        2.0       535.50   \n",
       "2011-03-31      0.000000             10.0        3.0      1089.01   \n",
       "2011-03-31      0.401786             20.0        7.0       986.33   \n",
       "\n",
       "            avg_revenue_90d  trans_date  \n",
       "dateindex                                \n",
       "2011-03-31       336.436667  2011-03-01  \n",
       "2011-03-31        79.978333  2011-03-25  \n",
       "2011-03-31       267.750000  2011-03-30  \n",
       "2011-03-31       363.003333  2011-03-07  \n",
       "2011-03-31       140.904286  2011-03-27  "
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "id_grouping.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "DatetimeIndex: 8055 entries, 2011-03-31 to 2011-12-31\n",
      "Data columns (total 12 columns):\n",
      "id                 8055 non-null float64\n",
      "invoices_no        8055 non-null int64\n",
      "categories         8055 non-null float64\n",
      "quantity           8055 non-null int64\n",
      "unitprice          8055 non-null float64\n",
      "gross              8055 non-null float64\n",
      "top_products       8055 non-null float64\n",
      "days_since_last    8055 non-null float64\n",
      "trans_90d          8055 non-null float64\n",
      "revenue_90d        8055 non-null float64\n",
      "avg_revenue_90d    8055 non-null float64\n",
      "trans_date         8055 non-null object\n",
      "dtypes: float64(9), int64(2), object(1)\n",
      "memory usage: 818.1+ KB\n"
     ]
    }
   ],
   "source": [
    "id_grouping.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2638"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "id_grouping.id.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "2638"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.id.nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "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>invoices_no</th>\n",
       "      <th>categories</th>\n",
       "      <th>quantity</th>\n",
       "      <th>unitprice</th>\n",
       "      <th>gross</th>\n",
       "      <th>top_products</th>\n",
       "      <th>days_since_last</th>\n",
       "      <th>trans_90d</th>\n",
       "      <th>revenue_90d</th>\n",
       "      <th>avg_revenue_90d</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>8055.000000</td>\n",
       "      <td>8055.000000</td>\n",
       "      <td>8055.000000</td>\n",
       "      <td>8055.000000</td>\n",
       "      <td>8055.000000</td>\n",
       "      <td>8055.000000</td>\n",
       "      <td>8055.000000</td>\n",
       "      <td>8055.000000</td>\n",
       "      <td>8055.000000</td>\n",
       "      <td>8055.000000</td>\n",
       "      <td>8055.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>15504.623712</td>\n",
       "      <td>1.685040</td>\n",
       "      <td>20.997726</td>\n",
       "      <td>343.505028</td>\n",
       "      <td>4.577203</td>\n",
       "      <td>583.294929</td>\n",
       "      <td>0.677068</td>\n",
       "      <td>49.804593</td>\n",
       "      <td>2.905897</td>\n",
       "      <td>1158.723470</td>\n",
       "      <td>299.430843</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1588.078860</td>\n",
       "      <td>1.508438</td>\n",
       "      <td>24.052673</td>\n",
       "      <td>763.548120</td>\n",
       "      <td>24.879742</td>\n",
       "      <td>1660.222234</td>\n",
       "      <td>0.246324</td>\n",
       "      <td>57.775746</td>\n",
       "      <td>4.166611</td>\n",
       "      <td>4068.858256</td>\n",
       "      <td>385.406778</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>12747.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-1350.000000</td>\n",
       "      <td>0.060000</td>\n",
       "      <td>-1000.370000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-1241.430000</td>\n",
       "      <td>-1241.430000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>14135.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>7.000000</td>\n",
       "      <td>86.000000</td>\n",
       "      <td>2.025500</td>\n",
       "      <td>173.450000</td>\n",
       "      <td>0.522727</td>\n",
       "      <td>11.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>165.940000</td>\n",
       "      <td>109.025000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>15498.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>15.000000</td>\n",
       "      <td>188.000000</td>\n",
       "      <td>2.799600</td>\n",
       "      <td>323.740000</td>\n",
       "      <td>0.697508</td>\n",
       "      <td>29.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>470.730000</td>\n",
       "      <td>226.842500</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>16873.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>26.000000</td>\n",
       "      <td>358.000000</td>\n",
       "      <td>3.950000</td>\n",
       "      <td>578.475000</td>\n",
       "      <td>0.875000</td>\n",
       "      <td>65.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>1015.975000</td>\n",
       "      <td>372.785000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>18287.000000</td>\n",
       "      <td>48.000000</td>\n",
       "      <td>530.000000</td>\n",
       "      <td>25662.000000</td>\n",
       "      <td>951.390000</td>\n",
       "      <td>70246.500000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>366.000000</td>\n",
       "      <td>84.000000</td>\n",
       "      <td>123737.110000</td>\n",
       "      <td>7152.500000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 id  invoices_no   categories      quantity    unitprice  \\\n",
       "count   8055.000000  8055.000000  8055.000000   8055.000000  8055.000000   \n",
       "mean   15504.623712     1.685040    20.997726    343.505028     4.577203   \n",
       "std     1588.078860     1.508438    24.052673    763.548120    24.879742   \n",
       "min    12747.000000     1.000000     1.000000  -1350.000000     0.060000   \n",
       "25%    14135.000000     1.000000     7.000000     86.000000     2.025500   \n",
       "50%    15498.000000     1.000000    15.000000    188.000000     2.799600   \n",
       "75%    16873.000000     2.000000    26.000000    358.000000     3.950000   \n",
       "max    18287.000000    48.000000   530.000000  25662.000000   951.390000   \n",
       "\n",
       "              gross  top_products  days_since_last    trans_90d  \\\n",
       "count   8055.000000   8055.000000      8055.000000  8055.000000   \n",
       "mean     583.294929      0.677068        49.804593     2.905897   \n",
       "std     1660.222234      0.246324        57.775746     4.166611   \n",
       "min    -1000.370000      0.000000         1.000000     0.000000   \n",
       "25%      173.450000      0.522727        11.000000     1.000000   \n",
       "50%      323.740000      0.697508        29.000000     2.000000   \n",
       "75%      578.475000      0.875000        65.000000     4.000000   \n",
       "max    70246.500000      1.000000       366.000000    84.000000   \n",
       "\n",
       "         revenue_90d  avg_revenue_90d  \n",
       "count    8055.000000      8055.000000  \n",
       "mean     1158.723470       299.430843  \n",
       "std      4068.858256       385.406778  \n",
       "min     -1241.430000     -1241.430000  \n",
       "25%       165.940000       109.025000  \n",
       "50%       470.730000       226.842500  \n",
       "75%      1015.975000       372.785000  \n",
       "max    123737.110000      7152.500000  "
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "id_grouping.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "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>invoices_no</th>\n",
       "      <th>categories</th>\n",
       "      <th>quantity</th>\n",
       "      <th>unitprice</th>\n",
       "      <th>gross</th>\n",
       "      <th>top_products</th>\n",
       "      <th>days_since_last</th>\n",
       "      <th>trans_90d</th>\n",
       "      <th>revenue_90d</th>\n",
       "      <th>avg_revenue_90d</th>\n",
       "      <th>cust_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>11706.000000</td>\n",
       "      <td>11706.000000</td>\n",
       "      <td>11706.000000</td>\n",
       "      <td>11706.000000</td>\n",
       "      <td>11706.000000</td>\n",
       "      <td>11706.000000</td>\n",
       "      <td>11706.000000</td>\n",
       "      <td>11706.000000</td>\n",
       "      <td>11706.000000</td>\n",
       "      <td>11706.000000</td>\n",
       "      <td>11706.000000</td>\n",
       "      <td>11706.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>15467.320178</td>\n",
       "      <td>1.159491</td>\n",
       "      <td>20.577055</td>\n",
       "      <td>236.368785</td>\n",
       "      <td>4.847908</td>\n",
       "      <td>401.370293</td>\n",
       "      <td>0.677865</td>\n",
       "      <td>42.643943</td>\n",
       "      <td>4.614215</td>\n",
       "      <td>1794.355276</td>\n",
       "      <td>326.865368</td>\n",
       "      <td>15467.320178</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>1620.324488</td>\n",
       "      <td>0.488853</td>\n",
       "      <td>27.504850</td>\n",
       "      <td>528.244587</td>\n",
       "      <td>33.150417</td>\n",
       "      <td>1125.771513</td>\n",
       "      <td>0.273778</td>\n",
       "      <td>53.687268</td>\n",
       "      <td>7.304577</td>\n",
       "      <td>5202.351173</td>\n",
       "      <td>451.680598</td>\n",
       "      <td>1620.324488</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>12747.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-3679.000000</td>\n",
       "      <td>0.010000</td>\n",
       "      <td>-7460.950000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>-1425.790000</td>\n",
       "      <td>-1241.430000</td>\n",
       "      <td>12747.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>14068.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>4.000000</td>\n",
       "      <td>48.000000</td>\n",
       "      <td>1.958022</td>\n",
       "      <td>113.790000</td>\n",
       "      <td>0.500000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>243.770000</td>\n",
       "      <td>123.773125</td>\n",
       "      <td>14068.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>15401.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>14.000000</td>\n",
       "      <td>138.000000</td>\n",
       "      <td>2.745000</td>\n",
       "      <td>275.535000</td>\n",
       "      <td>0.705882</td>\n",
       "      <td>22.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>627.015000</td>\n",
       "      <td>238.304091</td>\n",
       "      <td>15401.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>16869.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>26.000000</td>\n",
       "      <td>272.000000</td>\n",
       "      <td>3.948332</td>\n",
       "      <td>434.975000</td>\n",
       "      <td>0.911111</td>\n",
       "      <td>54.000000</td>\n",
       "      <td>5.000000</td>\n",
       "      <td>1458.380000</td>\n",
       "      <td>387.837857</td>\n",
       "      <td>16869.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>18287.000000</td>\n",
       "      <td>12.000000</td>\n",
       "      <td>543.000000</td>\n",
       "      <td>25930.000000</td>\n",
       "      <td>2500.000000</td>\n",
       "      <td>71601.440000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>366.000000</td>\n",
       "      <td>84.000000</td>\n",
       "      <td>123737.110000</td>\n",
       "      <td>9465.066667</td>\n",
       "      <td>18287.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 id   invoices_no    categories      quantity     unitprice  \\\n",
       "count  11706.000000  11706.000000  11706.000000  11706.000000  11706.000000   \n",
       "mean   15467.320178      1.159491     20.577055    236.368785      4.847908   \n",
       "std     1620.324488      0.488853     27.504850    528.244587     33.150417   \n",
       "min    12747.000000      1.000000      1.000000  -3679.000000      0.010000   \n",
       "25%    14068.000000      1.000000      4.000000     48.000000      1.958022   \n",
       "50%    15401.000000      1.000000     14.000000    138.000000      2.745000   \n",
       "75%    16869.000000      1.000000     26.000000    272.000000      3.948332   \n",
       "max    18287.000000     12.000000    543.000000  25930.000000   2500.000000   \n",
       "\n",
       "              gross  top_products  days_since_last     trans_90d  \\\n",
       "count  11706.000000  11706.000000     11706.000000  11706.000000   \n",
       "mean     401.370293      0.677865        42.643943      4.614215   \n",
       "std     1125.771513      0.273778        53.687268      7.304577   \n",
       "min    -7460.950000      0.000000         1.000000      0.000000   \n",
       "25%      113.790000      0.500000         8.000000      1.000000   \n",
       "50%      275.535000      0.705882        22.000000      2.000000   \n",
       "75%      434.975000      0.911111        54.000000      5.000000   \n",
       "max    71601.440000      1.000000       366.000000     84.000000   \n",
       "\n",
       "         revenue_90d  avg_revenue_90d       cust_id  \n",
       "count   11706.000000     11706.000000  11706.000000  \n",
       "mean     1794.355276       326.865368  15467.320178  \n",
       "std      5202.351173       451.680598   1620.324488  \n",
       "min     -1425.790000     -1241.430000  12747.000000  \n",
       "25%       243.770000       123.773125  14068.000000  \n",
       "50%       627.015000       238.304091  15401.000000  \n",
       "75%      1458.380000       387.837857  16869.000000  \n",
       "max    123737.110000      9465.066667  18287.000000  "
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The ratio of the monthly mean \"gross\" to the daily mean is 1.45. The ratio of the monthly mean \"quantity\" to the daily mean is also 1.45. These are also consistent with the ratio of the monthly mean \"invoices_no\" to the daily mean. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.DataFrame(id_grouping)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Deleting outliers from monthly data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "67"
      ]
     },
     "execution_count": 44,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 67 observations with \"gross\" of more than 3 std above the mean, but none of such magnitude below the mean\n",
    "df.gross[(df.gross > 583.29 + 3*1660.22)].count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "119"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 119 observations with \"gross\" of more than 3 std above the mean, but none of such magnitude below the mean\n",
    "df.invoices_no[(df.invoices_no > 1.68 + 3*1.51)].count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(8055, 12)"
      ]
     },
     "execution_count": 46,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(7892, 12)"
      ]
     },
     "execution_count": 47,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Deleting these outliers (163 observations in total), or 2% of the observations\n",
    "df = df[(df.gross < 583.29 + 3*1660.22)]\n",
    "df = df[(df.invoices_no < 1.68 + 3*1.51)]\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Customer clustering"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {},
   "outputs": [],
   "source": [
    "# For frequency and monetary, the higher the values the better the ranking, unlike recency\n",
    "def customer_cluster(cluster, group, df, ascending=True):\n",
    "    initial = (df.groupby(cluster)[group] # This step is to generate a sorted index based on the 'group' mean\n",
    "                .mean()\n",
    "                .reset_index()\n",
    "                .sort_values(by=group, ascending=True)\n",
    "                .reset_index(drop=True)) \n",
    "    initial['index'] = initial.index # Throwing out the old index as a scoring variable\n",
    "    final = df.reset_index().merge(initial, on=cluster).set_index('dateindex') # Merging with the original DF\n",
    "    final = (final.drop([cluster], axis=1)\n",
    "                .rename(columns={\"index\":cluster})) # Dropping the cluster scores, and renaming the old index as the cluster\n",
    "    return final"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {},
   "outputs": [],
   "source": [
    "def alt_cluster(cluster, group, df, ascending=False):\n",
    "    initial = (df.groupby(cluster)[group] # This step is to generate a sorted index based on the 'group' mean\n",
    "                .mean()\n",
    "                .reset_index()\n",
    "                .sort_values(by=group, ascending=False)\n",
    "                .reset_index(drop=True)) \n",
    "    initial['index'] = initial.index # Throwing out the old index as a scoring variable\n",
    "    final = df.reset_index().merge(initial, on=cluster).set_index('dateindex') # Merging with the original DF\n",
    "    final = (final.drop([cluster], axis=1)\n",
    "                .rename(columns={\"index\":cluster})) # Dropping the cluster scores, and renaming the old index as the cluster\n",
    "    return final"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [],
   "source": [
    "kmeans = KMeans(n_clusters=5) # Try for more granularity with 6 clusters"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 51,
   "metadata": {},
   "outputs": [],
   "source": [
    "kmeans.fit(df[['trans_90d']])\n",
    "df['frequency'] = kmeans.predict(df[['trans_90d']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 52,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = customer_cluster('frequency', 'trans_90d', df, True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 53,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(7892, 14)"
      ]
     },
     "execution_count": 53,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 54,
   "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>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>frequency</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>0</th>\n",
       "      <td>3654.0</td>\n",
       "      <td>0.626710</td>\n",
       "      <td>0.483744</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>1.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2919.0</td>\n",
       "      <td>2.725591</td>\n",
       "      <td>0.782169</td>\n",
       "      <td>2.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>852.0</td>\n",
       "      <td>5.715962</td>\n",
       "      <td>0.781140</td>\n",
       "      <td>5.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>7.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>396.0</td>\n",
       "      <td>9.636364</td>\n",
       "      <td>1.547261</td>\n",
       "      <td>8.0</td>\n",
       "      <td>8.0</td>\n",
       "      <td>9.0</td>\n",
       "      <td>11.0</td>\n",
       "      <td>13.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>71.0</td>\n",
       "      <td>18.323944</td>\n",
       "      <td>6.283254</td>\n",
       "      <td>14.0</td>\n",
       "      <td>15.0</td>\n",
       "      <td>16.0</td>\n",
       "      <td>20.0</td>\n",
       "      <td>57.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            count       mean       std   min   25%   50%   75%   max\n",
       "frequency                                                           \n",
       "0          3654.0   0.626710  0.483744   0.0   0.0   1.0   1.0   1.0\n",
       "1          2919.0   2.725591  0.782169   2.0   2.0   3.0   3.0   4.0\n",
       "2           852.0   5.715962  0.781140   5.0   5.0   6.0   6.0   7.0\n",
       "3           396.0   9.636364  1.547261   8.0   8.0   9.0  11.0  13.0\n",
       "4            71.0  18.323944  6.283254  14.0  15.0  16.0  20.0  57.0"
      ]
     },
     "execution_count": 54,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 46% of observations are in frequency score==0!\n",
    "df.groupby('frequency')['trans_90d_x'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 55,
   "metadata": {},
   "outputs": [],
   "source": [
    "kmeans.fit(df[['revenue_90d']])\n",
    "df['monetary'] = kmeans.predict(df[['revenue_90d']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = customer_cluster('monetary', 'revenue_90d', df, True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 57,
   "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>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>monetary</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>0</th>\n",
       "      <td>6308.0</td>\n",
       "      <td>383.472302</td>\n",
       "      <td>330.126232</td>\n",
       "      <td>-1241.43</td>\n",
       "      <td>87.795</td>\n",
       "      <td>325.92</td>\n",
       "      <td>622.510</td>\n",
       "      <td>1156.18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1359.0</td>\n",
       "      <td>1936.671127</td>\n",
       "      <td>682.983940</td>\n",
       "      <td>1157.89</td>\n",
       "      <td>1388.460</td>\n",
       "      <td>1742.72</td>\n",
       "      <td>2277.300</td>\n",
       "      <td>4001.63</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>187.0</td>\n",
       "      <td>6095.325989</td>\n",
       "      <td>1800.672933</td>\n",
       "      <td>4034.05</td>\n",
       "      <td>4629.190</td>\n",
       "      <td>5493.88</td>\n",
       "      <td>7093.870</td>\n",
       "      <td>11103.20</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>35.0</td>\n",
       "      <td>16378.442000</td>\n",
       "      <td>5162.500836</td>\n",
       "      <td>11365.02</td>\n",
       "      <td>13162.895</td>\n",
       "      <td>15107.09</td>\n",
       "      <td>17329.815</td>\n",
       "      <td>38391.26</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>3.0</td>\n",
       "      <td>90043.283333</td>\n",
       "      <td>11131.324294</td>\n",
       "      <td>80014.34</td>\n",
       "      <td>84054.970</td>\n",
       "      <td>88095.60</td>\n",
       "      <td>95057.755</td>\n",
       "      <td>102019.91</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           count          mean           std       min        25%       50%  \\\n",
       "monetary                                                                      \n",
       "0         6308.0    383.472302    330.126232  -1241.43     87.795    325.92   \n",
       "1         1359.0   1936.671127    682.983940   1157.89   1388.460   1742.72   \n",
       "2          187.0   6095.325989   1800.672933   4034.05   4629.190   5493.88   \n",
       "3           35.0  16378.442000   5162.500836  11365.02  13162.895  15107.09   \n",
       "4            3.0  90043.283333  11131.324294  80014.34  84054.970  88095.60   \n",
       "\n",
       "                75%        max  \n",
       "monetary                        \n",
       "0           622.510    1156.18  \n",
       "1          2277.300    4001.63  \n",
       "2          7093.870   11103.20  \n",
       "3         17329.815   38391.26  \n",
       "4         95057.755  102019.91  "
      ]
     },
     "execution_count": 57,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 80% of observations are in monetary score==0!\n",
    "df.groupby('monetary')['revenue_90d_x'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 58,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Doing the alternative monetary score, using the average gross instead of the sum\n",
    "kmeans.fit(df[['avg_revenue_90d']])\n",
    "df['monetary_avg'] = kmeans.predict(df[['avg_revenue_90d']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 59,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = customer_cluster('monetary_avg', 'avg_revenue_90d', df, True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 60,
   "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>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>monetary_avg</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>0</th>\n",
       "      <td>3069.0</td>\n",
       "      <td>60.886323</td>\n",
       "      <td>75.327583</td>\n",
       "      <td>-1241.430</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>46.046667</td>\n",
       "      <td>130.400000</td>\n",
       "      <td>172.75</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>3378.0</td>\n",
       "      <td>285.214969</td>\n",
       "      <td>70.531620</td>\n",
       "      <td>172.910</td>\n",
       "      <td>222.552083</td>\n",
       "      <td>282.003125</td>\n",
       "      <td>338.995833</td>\n",
       "      <td>433.70</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1141.0</td>\n",
       "      <td>583.002192</td>\n",
       "      <td>118.133768</td>\n",
       "      <td>434.015</td>\n",
       "      <td>481.380000</td>\n",
       "      <td>554.227500</td>\n",
       "      <td>664.510000</td>\n",
       "      <td>874.55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>262.0</td>\n",
       "      <td>1172.573069</td>\n",
       "      <td>256.988235</td>\n",
       "      <td>879.300</td>\n",
       "      <td>968.346250</td>\n",
       "      <td>1079.185000</td>\n",
       "      <td>1307.509250</td>\n",
       "      <td>1932.24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>42.0</td>\n",
       "      <td>2763.767055</td>\n",
       "      <td>751.706339</td>\n",
       "      <td>1974.132</td>\n",
       "      <td>2146.692500</td>\n",
       "      <td>2499.216429</td>\n",
       "      <td>3197.072500</td>\n",
       "      <td>4730.00</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               count         mean         std       min          25%  \\\n",
       "monetary_avg                                                           \n",
       "0             3069.0    60.886323   75.327583 -1241.430     0.000000   \n",
       "1             3378.0   285.214969   70.531620   172.910   222.552083   \n",
       "2             1141.0   583.002192  118.133768   434.015   481.380000   \n",
       "3              262.0  1172.573069  256.988235   879.300   968.346250   \n",
       "4               42.0  2763.767055  751.706339  1974.132  2146.692500   \n",
       "\n",
       "                      50%          75%      max  \n",
       "monetary_avg                                     \n",
       "0               46.046667   130.400000   172.75  \n",
       "1              282.003125   338.995833   433.70  \n",
       "2              554.227500   664.510000   874.55  \n",
       "3             1079.185000  1307.509250  1932.24  \n",
       "4             2499.216429  3197.072500  4730.00  "
      ]
     },
     "execution_count": 60,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Alternative monetary score is skewed, but much less than the initial one\n",
    "df.groupby('monetary_avg')['avg_revenue_90d_x'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [],
   "source": [
    "kmeans.fit(df[['days_since_last']])\n",
    "df['recency'] = kmeans.predict(df[['days_since_last']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = alt_cluster('recency', 'days_since_last', df, ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 63,
   "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>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>recency</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>0</th>\n",
       "      <td>174.0</td>\n",
       "      <td>283.936782</td>\n",
       "      <td>39.833490</td>\n",
       "      <td>232.0</td>\n",
       "      <td>250.0</td>\n",
       "      <td>275.0</td>\n",
       "      <td>315.0</td>\n",
       "      <td>366.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>433.0</td>\n",
       "      <td>178.101617</td>\n",
       "      <td>24.499316</td>\n",
       "      <td>142.0</td>\n",
       "      <td>158.0</td>\n",
       "      <td>174.0</td>\n",
       "      <td>198.0</td>\n",
       "      <td>231.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>1042.0</td>\n",
       "      <td>104.051823</td>\n",
       "      <td>17.764853</td>\n",
       "      <td>78.0</td>\n",
       "      <td>89.0</td>\n",
       "      <td>102.0</td>\n",
       "      <td>118.0</td>\n",
       "      <td>141.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2121.0</td>\n",
       "      <td>50.816124</td>\n",
       "      <td>12.714282</td>\n",
       "      <td>33.0</td>\n",
       "      <td>40.0</td>\n",
       "      <td>49.0</td>\n",
       "      <td>61.0</td>\n",
       "      <td>77.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4122.0</td>\n",
       "      <td>13.902475</td>\n",
       "      <td>8.855616</td>\n",
       "      <td>1.0</td>\n",
       "      <td>6.0</td>\n",
       "      <td>13.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>32.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          count        mean        std    min    25%    50%    75%    max\n",
       "recency                                                                  \n",
       "0         174.0  283.936782  39.833490  232.0  250.0  275.0  315.0  366.0\n",
       "1         433.0  178.101617  24.499316  142.0  158.0  174.0  198.0  231.0\n",
       "2        1042.0  104.051823  17.764853   78.0   89.0  102.0  118.0  141.0\n",
       "3        2121.0   50.816124  12.714282   33.0   40.0   49.0   61.0   77.0\n",
       "4        4122.0   13.902475   8.855616    1.0    6.0   13.0   21.0   32.0"
      ]
     },
     "execution_count": 63,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# More than half of the observations are in recency score==5.\n",
    "df.groupby('recency')['days_since_last_x'].describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The individual RFM scores are highly skewed, with Monetary being the worst. Let's try to spread out the distribution of the scores by taking the natural log values of the Recency variable (as it has no negative values), and the cube root of the Frequency and Monetary variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 64,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 569,
       "width": 846
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "M = df[['invoices_no', 'categories', 'quantity', 'unitprice', 'gross', 'frequency', 'monetary', 'monetary_avg', 'recency']]\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(14, 8))\n",
    "\n",
    "mask = np.zeros_like(M.corr(), dtype=np.bool)\n",
    "mask[np.triu_indices_from(mask)] = True\n",
    "\n",
    "sns.heatmap(M.corr(), mask=mask, annot=True, ax=ax)\n",
    "ax.set_yticks(np.arange(0, df.shape[1])+0.1)\n",
    "ax.set_ylim([M.shape[1], 0])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 65,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['id', 'invoices_no', 'categories', 'quantity', 'unitprice', 'gross',\n",
       "       'top_products', 'days_since_last_x', 'trans_90d_x', 'revenue_90d_x',\n",
       "       'avg_revenue_90d_x', 'trans_date', 'trans_90d_y', 'frequency',\n",
       "       'revenue_90d_y', 'monetary', 'avg_revenue_90d_y', 'monetary_avg',\n",
       "       'days_since_last_y', 'recency'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 65,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 66,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.drop(['trans_90d_y', 'frequency', 'revenue_90d_y', 'monetary', 'avg_revenue_90d_y', 'monetary_avg',\n",
    "       'days_since_last_y', 'recency'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 67,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.columns = ['id', 'invoices_no', 'categories', 'quantity', 'unitprice', 'gross', 'top_products', 'days_since_last', \n",
    "              'trans_90d', 'revenue_90d', 'avg_revenue_90d', 'trans_date']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Clustering using transformed RFM indicators"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 68,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['ln_days_since_last'] = np.log(df.days_since_last)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 69,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['cube_trans_90d'] = np.cbrt(df.trans_90d)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 70,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['cube_revenue_90d'] = np.cbrt(df.revenue_90d)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 71,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['cube_avg_revenue_90d'] = np.cbrt(df.avg_revenue_90d)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 72,
   "metadata": {},
   "outputs": [],
   "source": [
    "kmeans.fit(df[['cube_trans_90d']])\n",
    "df['frequency'] = kmeans.predict(df[['cube_trans_90d']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 73,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = customer_cluster('frequency', 'cube_trans_90d', df, True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 74,
   "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>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>frequency</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>0</th>\n",
       "      <td>1364.0</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",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2290.0</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2317.0</td>\n",
       "      <td>1.331845</td>\n",
       "      <td>0.089130</td>\n",
       "      <td>1.259921</td>\n",
       "      <td>1.259921</td>\n",
       "      <td>1.259921</td>\n",
       "      <td>1.442250</td>\n",
       "      <td>1.442250</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1454.0</td>\n",
       "      <td>1.702828</td>\n",
       "      <td>0.114734</td>\n",
       "      <td>1.587401</td>\n",
       "      <td>1.587401</td>\n",
       "      <td>1.709976</td>\n",
       "      <td>1.817121</td>\n",
       "      <td>1.912931</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>467.0</td>\n",
       "      <td>2.196762</td>\n",
       "      <td>0.224454</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>2.154435</td>\n",
       "      <td>2.289428</td>\n",
       "      <td>3.848501</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            count      mean       std       min       25%       50%       75%  \\\n",
       "frequency                                                                       \n",
       "0          1364.0  0.000000  0.000000  0.000000  0.000000  0.000000  0.000000   \n",
       "1          2290.0  1.000000  0.000000  1.000000  1.000000  1.000000  1.000000   \n",
       "2          2317.0  1.331845  0.089130  1.259921  1.259921  1.259921  1.442250   \n",
       "3          1454.0  1.702828  0.114734  1.587401  1.587401  1.709976  1.817121   \n",
       "4           467.0  2.196762  0.224454  2.000000  2.000000  2.154435  2.289428   \n",
       "\n",
       "                max  \n",
       "frequency            \n",
       "0          0.000000  \n",
       "1          1.000000  \n",
       "2          1.442250  \n",
       "3          1.912931  \n",
       "4          3.848501  "
      ]
     },
     "execution_count": 74,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# The Frequency score's distribution is a lot less skewed now\n",
    "df.groupby('frequency')['cube_trans_90d_x'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 75,
   "metadata": {},
   "outputs": [],
   "source": [
    "kmeans.fit(df[['cube_revenue_90d']])\n",
    "df['monetary'] = kmeans.predict(df[['cube_revenue_90d']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 76,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = customer_cluster('monetary', 'cube_revenue_90d', df, True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 77,
   "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>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>monetary</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>0</th>\n",
       "      <td>1488.0</td>\n",
       "      <td>-0.133693</td>\n",
       "      <td>1.024753</td>\n",
       "      <td>-10.747499</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>3.149803</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2669.0</td>\n",
       "      <td>6.453296</td>\n",
       "      <td>0.984265</td>\n",
       "      <td>3.271066</td>\n",
       "      <td>5.770700</td>\n",
       "      <td>6.702872</td>\n",
       "      <td>7.221361</td>\n",
       "      <td>7.890166</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2492.0</td>\n",
       "      <td>9.351108</td>\n",
       "      <td>0.895683</td>\n",
       "      <td>7.893109</td>\n",
       "      <td>8.585232</td>\n",
       "      <td>9.251234</td>\n",
       "      <td>10.039097</td>\n",
       "      <td>11.157460</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1072.0</td>\n",
       "      <td>12.988237</td>\n",
       "      <td>1.425156</td>\n",
       "      <td>11.159280</td>\n",
       "      <td>11.844587</td>\n",
       "      <td>12.602936</td>\n",
       "      <td>13.840759</td>\n",
       "      <td>16.771024</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>171.0</td>\n",
       "      <td>20.572547</td>\n",
       "      <td>4.471375</td>\n",
       "      <td>16.786050</td>\n",
       "      <td>17.796825</td>\n",
       "      <td>19.210693</td>\n",
       "      <td>21.814398</td>\n",
       "      <td>46.726327</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           count       mean       std        min        25%        50%  \\\n",
       "monetary                                                                 \n",
       "0         1488.0  -0.133693  1.024753 -10.747499   0.000000   0.000000   \n",
       "1         2669.0   6.453296  0.984265   3.271066   5.770700   6.702872   \n",
       "2         2492.0   9.351108  0.895683   7.893109   8.585232   9.251234   \n",
       "3         1072.0  12.988237  1.425156  11.159280  11.844587  12.602936   \n",
       "4          171.0  20.572547  4.471375  16.786050  17.796825  19.210693   \n",
       "\n",
       "                75%        max  \n",
       "monetary                        \n",
       "0          0.000000   3.149803  \n",
       "1          7.221361   7.890166  \n",
       "2         10.039097  11.157460  \n",
       "3         13.840759  16.771024  \n",
       "4         21.814398  46.726327  "
      ]
     },
     "execution_count": 77,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# So too for Monetary\n",
    "df.groupby('monetary')['cube_revenue_90d_x'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 78,
   "metadata": {},
   "outputs": [],
   "source": [
    "kmeans.fit(df[['cube_avg_revenue_90d']])\n",
    "df['monetary_avg'] = kmeans.predict(df[['cube_avg_revenue_90d']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 79,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = customer_cluster('monetary_avg', 'cube_avg_revenue_90d', df, True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 80,
   "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>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>monetary_avg</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>0</th>\n",
       "      <td>1474.0</td>\n",
       "      <td>-0.148595</td>\n",
       "      <td>0.913349</td>\n",
       "      <td>-10.747499</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2.441299</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2017.0</td>\n",
       "      <td>5.056562</td>\n",
       "      <td>0.622358</td>\n",
       "      <td>2.458209</td>\n",
       "      <td>4.767195</td>\n",
       "      <td>5.199316</td>\n",
       "      <td>5.513027</td>\n",
       "      <td>5.831086</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2780.0</td>\n",
       "      <td>6.603201</td>\n",
       "      <td>0.441415</td>\n",
       "      <td>5.832236</td>\n",
       "      <td>6.217832</td>\n",
       "      <td>6.634719</td>\n",
       "      <td>6.954874</td>\n",
       "      <td>7.412069</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1323.0</td>\n",
       "      <td>8.212566</td>\n",
       "      <td>0.589751</td>\n",
       "      <td>7.413404</td>\n",
       "      <td>7.706606</td>\n",
       "      <td>8.090992</td>\n",
       "      <td>8.643555</td>\n",
       "      <td>9.599309</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>298.0</td>\n",
       "      <td>10.996038</td>\n",
       "      <td>1.439316</td>\n",
       "      <td>9.617056</td>\n",
       "      <td>9.952785</td>\n",
       "      <td>10.549156</td>\n",
       "      <td>11.522346</td>\n",
       "      <td>16.786251</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "               count       mean       std        min       25%        50%  \\\n",
       "monetary_avg                                                                \n",
       "0             1474.0  -0.148595  0.913349 -10.747499  0.000000   0.000000   \n",
       "1             2017.0   5.056562  0.622358   2.458209  4.767195   5.199316   \n",
       "2             2780.0   6.603201  0.441415   5.832236  6.217832   6.634719   \n",
       "3             1323.0   8.212566  0.589751   7.413404  7.706606   8.090992   \n",
       "4              298.0  10.996038  1.439316   9.617056  9.952785  10.549156   \n",
       "\n",
       "                    75%        max  \n",
       "monetary_avg                        \n",
       "0              0.000000   2.441299  \n",
       "1              5.513027   5.831086  \n",
       "2              6.954874   7.412069  \n",
       "3              8.643555   9.599309  \n",
       "4             11.522346  16.786251  "
      ]
     },
     "execution_count": 80,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby('monetary_avg')['cube_avg_revenue_90d_x'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 81,
   "metadata": {},
   "outputs": [],
   "source": [
    "kmeans.fit(df[['ln_days_since_last']])\n",
    "df['recency'] = kmeans.predict(df[['ln_days_since_last']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 82,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = alt_cluster('recency', 'ln_days_since_last', df, ascending=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 83,
   "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>count</th>\n",
       "      <th>mean</th>\n",
       "      <th>std</th>\n",
       "      <th>min</th>\n",
       "      <th>25%</th>\n",
       "      <th>50%</th>\n",
       "      <th>75%</th>\n",
       "      <th>max</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>recency</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>0</th>\n",
       "      <td>1440.0</td>\n",
       "      <td>4.948472</td>\n",
       "      <td>0.362411</td>\n",
       "      <td>4.454347</td>\n",
       "      <td>4.651568</td>\n",
       "      <td>4.867534</td>\n",
       "      <td>5.187386</td>\n",
       "      <td>5.902633</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2330.0</td>\n",
       "      <td>3.942678</td>\n",
       "      <td>0.278101</td>\n",
       "      <td>3.496508</td>\n",
       "      <td>3.713572</td>\n",
       "      <td>3.912023</td>\n",
       "      <td>4.174387</td>\n",
       "      <td>4.442651</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2115.0</td>\n",
       "      <td>3.016628</td>\n",
       "      <td>0.277186</td>\n",
       "      <td>2.564949</td>\n",
       "      <td>2.772589</td>\n",
       "      <td>3.044522</td>\n",
       "      <td>3.295837</td>\n",
       "      <td>3.465736</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>1525.0</td>\n",
       "      <td>1.964741</td>\n",
       "      <td>0.334978</td>\n",
       "      <td>1.386294</td>\n",
       "      <td>1.791759</td>\n",
       "      <td>1.945910</td>\n",
       "      <td>2.197225</td>\n",
       "      <td>2.484907</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>482.0</td>\n",
       "      <td>0.590461</td>\n",
       "      <td>0.457092</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>0.693147</td>\n",
       "      <td>1.098612</td>\n",
       "      <td>1.098612</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          count      mean       std       min       25%       50%       75%  \\\n",
       "recency                                                                       \n",
       "0        1440.0  4.948472  0.362411  4.454347  4.651568  4.867534  5.187386   \n",
       "1        2330.0  3.942678  0.278101  3.496508  3.713572  3.912023  4.174387   \n",
       "2        2115.0  3.016628  0.277186  2.564949  2.772589  3.044522  3.295837   \n",
       "3        1525.0  1.964741  0.334978  1.386294  1.791759  1.945910  2.197225   \n",
       "4         482.0  0.590461  0.457092  0.000000  0.000000  0.693147  1.098612   \n",
       "\n",
       "              max  \n",
       "recency            \n",
       "0        5.902633  \n",
       "1        4.442651  \n",
       "2        3.465736  \n",
       "3        2.484907  \n",
       "4        1.098612  "
      ]
     },
     "execution_count": 83,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby('recency')['ln_days_since_last_x'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 84,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['id', 'invoices_no', 'categories', 'quantity', 'unitprice', 'gross',\n",
       "       'top_products', 'days_since_last', 'trans_90d', 'revenue_90d',\n",
       "       'avg_revenue_90d', 'trans_date', 'ln_days_since_last_x',\n",
       "       'cube_trans_90d_x', 'cube_revenue_90d_x', 'cube_avg_revenue_90d_x',\n",
       "       'cube_trans_90d_y', 'frequency', 'cube_revenue_90d_y', 'monetary',\n",
       "       'cube_avg_revenue_90d_y', 'monetary_avg', 'ln_days_since_last_y',\n",
       "       'recency'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 84,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 85,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.drop(['ln_days_since_last_x', 'cube_trans_90d_x', 'cube_revenue_90d_x', 'cube_avg_revenue_90d_x', 'cube_trans_90d_y', \n",
    "         'cube_revenue_90d_y', 'cube_avg_revenue_90d_y', 'ln_days_since_last_y',\n",
    "         'top_products'], axis=1, inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 86,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>invoices_no</th>\n",
       "      <th>categories</th>\n",
       "      <th>quantity</th>\n",
       "      <th>unitprice</th>\n",
       "      <th>gross</th>\n",
       "      <th>days_since_last</th>\n",
       "      <th>trans_90d</th>\n",
       "      <th>revenue_90d</th>\n",
       "      <th>avg_revenue_90d</th>\n",
       "      <th>trans_date</th>\n",
       "      <th>frequency</th>\n",
       "      <th>monetary</th>\n",
       "      <th>monetary_avg</th>\n",
       "      <th>recency</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>dateindex</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>2011-03-31</th>\n",
       "      <td>12747.0</td>\n",
       "      <td>1</td>\n",
       "      <td>12.0</td>\n",
       "      <td>146</td>\n",
       "      <td>3.106667</td>\n",
       "      <td>310.78</td>\n",
       "      <td>40.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1009.31</td>\n",
       "      <td>336.436667</td>\n",
       "      <td>2011-03-01</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-03-31</th>\n",
       "      <td>16011.0</td>\n",
       "      <td>1</td>\n",
       "      <td>8.0</td>\n",
       "      <td>63</td>\n",
       "      <td>4.025000</td>\n",
       "      <td>118.55</td>\n",
       "      <td>35.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>347.09</td>\n",
       "      <td>173.545000</td>\n",
       "      <td>2011-03-29</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-03-31</th>\n",
       "      <td>14194.0</td>\n",
       "      <td>5</td>\n",
       "      <td>24.0</td>\n",
       "      <td>628</td>\n",
       "      <td>3.923021</td>\n",
       "      <td>1023.04</td>\n",
       "      <td>2.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1546.74</td>\n",
       "      <td>309.348000</td>\n",
       "      <td>2011-03-27</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-03-31</th>\n",
       "      <td>15159.0</td>\n",
       "      <td>3</td>\n",
       "      <td>6.5</td>\n",
       "      <td>131</td>\n",
       "      <td>5.245833</td>\n",
       "      <td>636.25</td>\n",
       "      <td>3.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>1491.24</td>\n",
       "      <td>372.810000</td>\n",
       "      <td>2011-03-17</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-03-31</th>\n",
       "      <td>16670.0</td>\n",
       "      <td>2</td>\n",
       "      <td>21.0</td>\n",
       "      <td>356</td>\n",
       "      <td>2.987875</td>\n",
       "      <td>582.48</td>\n",
       "      <td>8.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>1387.94</td>\n",
       "      <td>277.588000</td>\n",
       "      <td>2011-03-10</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-31</th>\n",
       "      <td>14141.0</td>\n",
       "      <td>1</td>\n",
       "      <td>10.0</td>\n",
       "      <td>77</td>\n",
       "      <td>3.842000</td>\n",
       "      <td>226.35</td>\n",
       "      <td>57.0</td>\n",
       "      <td>4.0</td>\n",
       "      <td>367.52</td>\n",
       "      <td>91.880000</td>\n",
       "      <td>2011-12-07</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-31</th>\n",
       "      <td>14621.0</td>\n",
       "      <td>1</td>\n",
       "      <td>16.0</td>\n",
       "      <td>254</td>\n",
       "      <td>1.702500</td>\n",
       "      <td>302.86</td>\n",
       "      <td>29.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>482.59</td>\n",
       "      <td>482.590000</td>\n",
       "      <td>2011-12-08</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-31</th>\n",
       "      <td>16711.0</td>\n",
       "      <td>1</td>\n",
       "      <td>5.0</td>\n",
       "      <td>344</td>\n",
       "      <td>1.112000</td>\n",
       "      <td>295.60</td>\n",
       "      <td>23.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>445.56</td>\n",
       "      <td>445.560000</td>\n",
       "      <td>2011-12-06</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-31</th>\n",
       "      <td>17659.0</td>\n",
       "      <td>1</td>\n",
       "      <td>2.0</td>\n",
       "      <td>-3</td>\n",
       "      <td>6.015000</td>\n",
       "      <td>-14.11</td>\n",
       "      <td>14.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>1346.43</td>\n",
       "      <td>448.810000</td>\n",
       "      <td>2011-12-06</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2011-12-31</th>\n",
       "      <td>17884.0</td>\n",
       "      <td>1</td>\n",
       "      <td>24.0</td>\n",
       "      <td>299</td>\n",
       "      <td>1.182083</td>\n",
       "      <td>114.61</td>\n",
       "      <td>308.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.00</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2011-12-06</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>7892 rows × 15 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                 id  invoices_no  categories  quantity  unitprice    gross  \\\n",
       "dateindex                                                                    \n",
       "2011-03-31  12747.0            1        12.0       146   3.106667   310.78   \n",
       "2011-03-31  16011.0            1         8.0        63   4.025000   118.55   \n",
       "2011-03-31  14194.0            5        24.0       628   3.923021  1023.04   \n",
       "2011-03-31  15159.0            3         6.5       131   5.245833   636.25   \n",
       "2011-03-31  16670.0            2        21.0       356   2.987875   582.48   \n",
       "...             ...          ...         ...       ...        ...      ...   \n",
       "2011-12-31  14141.0            1        10.0        77   3.842000   226.35   \n",
       "2011-12-31  14621.0            1        16.0       254   1.702500   302.86   \n",
       "2011-12-31  16711.0            1         5.0       344   1.112000   295.60   \n",
       "2011-12-31  17659.0            1         2.0        -3   6.015000   -14.11   \n",
       "2011-12-31  17884.0            1        24.0       299   1.182083   114.61   \n",
       "\n",
       "            days_since_last  trans_90d  revenue_90d  avg_revenue_90d  \\\n",
       "dateindex                                                              \n",
       "2011-03-31             40.0        3.0      1009.31       336.436667   \n",
       "2011-03-31             35.0        2.0       347.09       173.545000   \n",
       "2011-03-31              2.0        5.0      1546.74       309.348000   \n",
       "2011-03-31              3.0        4.0      1491.24       372.810000   \n",
       "2011-03-31              8.0        5.0      1387.94       277.588000   \n",
       "...                     ...        ...          ...              ...   \n",
       "2011-12-31             57.0        4.0       367.52        91.880000   \n",
       "2011-12-31             29.0        1.0       482.59       482.590000   \n",
       "2011-12-31             23.0        1.0       445.56       445.560000   \n",
       "2011-12-31             14.0        3.0      1346.43       448.810000   \n",
       "2011-12-31            308.0        0.0         0.00         0.000000   \n",
       "\n",
       "            trans_date  frequency  monetary  monetary_avg  recency  \n",
       "dateindex                                                           \n",
       "2011-03-31  2011-03-01          2         2             2        1  \n",
       "2011-03-31  2011-03-29          2         1             1        1  \n",
       "2011-03-31  2011-03-27          3         3             2        4  \n",
       "2011-03-31  2011-03-17          3         3             2        4  \n",
       "2011-03-31  2011-03-10          3         2             2        3  \n",
       "...                ...        ...       ...           ...      ...  \n",
       "2011-12-31  2011-12-07          3         1             1        1  \n",
       "2011-12-31  2011-12-08          1         1             3        2  \n",
       "2011-12-31  2011-12-06          1         1             3        2  \n",
       "2011-12-31  2011-12-06          2         2             3        2  \n",
       "2011-12-31  2011-12-06          0         0             0        0  \n",
       "\n",
       "[7892 rows x 15 columns]"
      ]
     },
     "execution_count": 86,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.sort_index()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Creating a composite equal-weighted RFM score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 87,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Using equal weights\n",
    "df['rfm_equal'] = df['recency'] + df['frequency'] + df['monetary']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 88,
   "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>days_since_last</th>\n",
       "      <th>trans_90d</th>\n",
       "      <th>revenue_90d</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>rfm_equal</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>154.814516</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>9.405966e-16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>87.800000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>-6.490000e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>74.504673</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>8.209991e+01</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>58.163471</td>\n",
       "      <td>1.022200</td>\n",
       "      <td>2.544462e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>42.089234</td>\n",
       "      <td>1.497575</td>\n",
       "      <td>3.881496e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>29.225259</td>\n",
       "      <td>1.897267</td>\n",
       "      <td>5.646622e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>20.217259</td>\n",
       "      <td>2.504569</td>\n",
       "      <td>7.578820e+02</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>15.806122</td>\n",
       "      <td>3.667800</td>\n",
       "      <td>1.030453e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>11.948795</td>\n",
       "      <td>4.701807</td>\n",
       "      <td>1.552567e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>9.328358</td>\n",
       "      <td>6.502488</td>\n",
       "      <td>2.300899e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>7.506073</td>\n",
       "      <td>8.947368</td>\n",
       "      <td>3.660647e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>4.508621</td>\n",
       "      <td>12.008621</td>\n",
       "      <td>7.754245e+03</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>1.781250</td>\n",
       "      <td>15.781250</td>\n",
       "      <td>1.123714e+04</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           days_since_last  trans_90d   revenue_90d\n",
       "rfm_equal                                          \n",
       "0               154.814516   0.000000  9.405966e-16\n",
       "1                87.800000   1.000000 -6.490000e+01\n",
       "2                74.504673   1.000000  8.209991e+01\n",
       "3                58.163471   1.022200  2.544462e+02\n",
       "4                42.089234   1.497575  3.881496e+02\n",
       "5                29.225259   1.897267  5.646622e+02\n",
       "6                20.217259   2.504569  7.578820e+02\n",
       "7                15.806122   3.667800  1.030453e+03\n",
       "8                11.948795   4.701807  1.552567e+03\n",
       "9                 9.328358   6.502488  2.300899e+03\n",
       "10                7.506073   8.947368  3.660647e+03\n",
       "11                4.508621  12.008621  7.754245e+03\n",
       "12                1.781250  15.781250  1.123714e+04"
      ]
     },
     "execution_count": 88,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby('rfm_equal')['days_since_last', 'trans_90d', 'revenue_90d'].mean()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 89,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 569,
       "width": 846
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# The transformed RFM variables show similar strong positive correlations to the resultant \"rfm_score\" and \"cust_class\"\n",
    "M = df[['invoices_no', 'categories', 'quantity', 'unitprice', 'gross', 'frequency', 'monetary', 'monetary_avg', \n",
    "        'recency', 'rfm_equal']]\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(14, 8))\n",
    "\n",
    "mask = np.zeros_like(M.corr(), dtype=np.bool)\n",
    "mask[np.triu_indices_from(mask)] = True\n",
    "\n",
    "sns.heatmap(M.corr(), mask=mask, annot=True, ax=ax)\n",
    "ax.set_yticks(np.arange(0, df.shape[1])+0.1)\n",
    "ax.set_ylim([M.shape[1], 0])\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "What's interesting about the correlation chart above is that the individual Recency, Frequency and Monetary indicators have generally weak relationships with \"gross\", though all are positive. The individual RFM scores have stronger correlations (all positive again) with \"invoices_no\" or the number of monthly transactions. This is likely due to the existence of infrequent but large buyers in the dataset. \n",
    "\n",
    "This leads to the issue of the RFM framework being unintentionally weighted in favour of regular customers, if the three scores are simply added across with equal weighting into a composite indicator, as in the \"rfm_score\" variable. We can observe this in the correlation of +0.56 between the \"rfm_equal\" and \"invoices_no\", but only +0.34 between \"rfm_equal\" and \"gross\". So a naive equally-weighted composite RFM indicator is largely measuring the number of transactions by a customer per month. This is fine, but what if we wanted the composite indicator to track other customer metrics?\n",
    "\n",
    "The \"monetary_avg\" score is also confirmed to be much less correlated to \"frequency\" than the initial \"monetary\" score. It would be better to use indicators that are not highly correlated to each other for crafting a composite indicator. As such, I would use the \"monetary_avg\" score from here as the Monetary indicator."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Considerations for constructing the RFM composite indicator"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There is no widely agreed methodology on how to weigh individual indicators for aggregation into a composite indicator. Weights typically have a major effect on the composite indicator value and on the resulting ranking, which is why weighting models need to be made explicit and transparent. \n",
    "\n",
    "Some commonly commonly used methods for weighting include the following:\n",
    "- Equal weights\n",
    "- Principal components analysis\n",
    "- Regression analysis \n",
    "- Theoretical considerations\n",
    "- Expert opinion\n",
    "\n",
    "The point of crafting specialised indicators implies eschewing equal weights, and there is no theory to guide us here. And the three individual indicators are taken directly from the RFM framework and is already limited, so I don't think we need to undertake PCA. I will instead pursue the regression analysis approach."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "For the example of how to go about building a bespoke composite RFM indicator, I would aim to build an indicator of monthly customer spending. That is after all the raison d'etre of businesses, though of course other aims are possible too. For this purpose, the monthly \"gross\" revenue will consequently be the dependent variable or target for the regression analysis."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 90,
   "metadata": {},
   "outputs": [],
   "source": [
    "#df.to_csv(r'C:\\\\Users\\\\Tan\\\\Desktop\\\\GA\\\\Capstone\\\\know_customer_monthly.csv', index=False)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}