RFM_Analysis_original.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Interactive shell of Ipython is used to execute multiple statements at a time\n",
    "from IPython.core.interactiveshell import InteractiveShell\n",
    "InteractiveShell.ast_node_interactivity='all'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import sklearn as skt\n",
    "import seaborn as sns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>CustomerKey</th>\n",
       "      <th>EnglishProductCategoryName</th>\n",
       "      <th>Model</th>\n",
       "      <th>Country</th>\n",
       "      <th>Region</th>\n",
       "      <th>Age</th>\n",
       "      <th>IncomeGroup</th>\n",
       "      <th>CalendarYear</th>\n",
       "      <th>OrderDate</th>\n",
       "      <th>OrderNumber</th>\n",
       "      <th>LineNumber</th>\n",
       "      <th>Quantity</th>\n",
       "      <th>Amount</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11000</td>\n",
       "      <td>Bikes</td>\n",
       "      <td>Mountain-200</td>\n",
       "      <td>Australia</td>\n",
       "      <td>Pacific</td>\n",
       "      <td>49</td>\n",
       "      <td>High</td>\n",
       "      <td>2013</td>\n",
       "      <td>2013-01-18</td>\n",
       "      <td>SO51522</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2319.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11000</td>\n",
       "      <td>Accessories</td>\n",
       "      <td>Fender Set - Mountain</td>\n",
       "      <td>Australia</td>\n",
       "      <td>Pacific</td>\n",
       "      <td>49</td>\n",
       "      <td>High</td>\n",
       "      <td>2013</td>\n",
       "      <td>2013-01-18</td>\n",
       "      <td>SO51522</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>21.98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11000</td>\n",
       "      <td>Bikes</td>\n",
       "      <td>Touring-1000</td>\n",
       "      <td>Australia</td>\n",
       "      <td>Pacific</td>\n",
       "      <td>49</td>\n",
       "      <td>High</td>\n",
       "      <td>2013</td>\n",
       "      <td>2013-05-03</td>\n",
       "      <td>SO57418</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2384.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>11000</td>\n",
       "      <td>Accessories</td>\n",
       "      <td>Touring Tire</td>\n",
       "      <td>Australia</td>\n",
       "      <td>Pacific</td>\n",
       "      <td>49</td>\n",
       "      <td>High</td>\n",
       "      <td>2013</td>\n",
       "      <td>2013-05-03</td>\n",
       "      <td>SO57418</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>28.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11000</td>\n",
       "      <td>Accessories</td>\n",
       "      <td>Touring Tire Tube</td>\n",
       "      <td>Australia</td>\n",
       "      <td>Pacific</td>\n",
       "      <td>49</td>\n",
       "      <td>High</td>\n",
       "      <td>2013</td>\n",
       "      <td>2013-05-03</td>\n",
       "      <td>SO57418</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>4.99</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CustomerKey EnglishProductCategoryName                           Model  \\\n",
       "0        11000       Bikes                 Mountain-200                     \n",
       "1        11000       Accessories           Fender Set - Mountain            \n",
       "2        11000       Bikes                 Touring-1000                     \n",
       "3        11000       Accessories           Touring Tire                     \n",
       "4        11000       Accessories           Touring Tire Tube                \n",
       "\n",
       "           Country           Region  Age IncomeGroup  CalendarYear  OrderDate  \\\n",
       "0  Australia        Pacific           49    High              2013 2013-01-18   \n",
       "1  Australia        Pacific           49    High              2013 2013-01-18   \n",
       "2  Australia        Pacific           49    High              2013 2013-05-03   \n",
       "3  Australia        Pacific           49    High              2013 2013-05-03   \n",
       "4  Australia        Pacific           49    High              2013 2013-05-03   \n",
       "\n",
       "  OrderNumber  LineNumber  Quantity   Amount  \n",
       "0  SO51522              1         1  2319.99  \n",
       "1  SO51522              2         1    21.98  \n",
       "2  SO57418              1         1  2384.07  \n",
       "3  SO57418              2         1    28.99  \n",
       "4  SO57418              3         1     4.99  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>CustomerKey</th>\n",
       "      <th>Age</th>\n",
       "      <th>CalendarYear</th>\n",
       "      <th>LineNumber</th>\n",
       "      <th>Quantity</th>\n",
       "      <th>Amount</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>54771.000000</td>\n",
       "      <td>54771.000000</td>\n",
       "      <td>54771.000000</td>\n",
       "      <td>54771.000000</td>\n",
       "      <td>54771.0</td>\n",
       "      <td>54771.00000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>18833.144894</td>\n",
       "      <td>51.304157</td>\n",
       "      <td>2013.035968</td>\n",
       "      <td>1.973216</td>\n",
       "      <td>1.0</td>\n",
       "      <td>299.37823</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>5407.123198</td>\n",
       "      <td>11.437086</td>\n",
       "      <td>0.186212</td>\n",
       "      <td>1.024153</td>\n",
       "      <td>0.0</td>\n",
       "      <td>677.39973</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>11000.000000</td>\n",
       "      <td>34.000000</td>\n",
       "      <td>2013.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2.29000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>14011.000000</td>\n",
       "      <td>42.000000</td>\n",
       "      <td>2013.000000</td>\n",
       "      <td>1.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>4.99000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>18176.000000</td>\n",
       "      <td>49.000000</td>\n",
       "      <td>2013.000000</td>\n",
       "      <td>2.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>24.99000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>23381.500000</td>\n",
       "      <td>59.000000</td>\n",
       "      <td>2013.000000</td>\n",
       "      <td>3.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>53.99000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>29480.000000</td>\n",
       "      <td>104.000000</td>\n",
       "      <td>2014.000000</td>\n",
       "      <td>8.000000</td>\n",
       "      <td>1.0</td>\n",
       "      <td>2443.35000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        CustomerKey           Age  CalendarYear    LineNumber  Quantity  \\\n",
       "count  54771.000000  54771.000000  54771.000000  54771.000000   54771.0   \n",
       "mean   18833.144894     51.304157   2013.035968      1.973216       1.0   \n",
       "std     5407.123198     11.437086      0.186212      1.024153       0.0   \n",
       "min    11000.000000     34.000000   2013.000000      1.000000       1.0   \n",
       "25%    14011.000000     42.000000   2013.000000      1.000000       1.0   \n",
       "50%    18176.000000     49.000000   2013.000000      2.000000       1.0   \n",
       "75%    23381.500000     59.000000   2013.000000      3.000000       1.0   \n",
       "max    29480.000000    104.000000   2014.000000      8.000000       1.0   \n",
       "\n",
       "            Amount  \n",
       "count  54771.00000  \n",
       "mean     299.37823  \n",
       "std      677.39973  \n",
       "min        2.29000  \n",
       "25%        4.99000  \n",
       "50%       24.99000  \n",
       "75%       53.99000  \n",
       "max     2443.35000  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aw_df = pd.read_excel('Adventure_Works_DB_2013.xlsx')\n",
    "aw_df.head()\n",
    "aw_df.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CustomerKey                   0\n",
       "EnglishProductCategoryName    0\n",
       "Model                         0\n",
       "Country                       0\n",
       "Region                        0\n",
       "Age                           0\n",
       "IncomeGroup                   0\n",
       "CalendarYear                  0\n",
       "OrderDate                     0\n",
       "OrderNumber                   0\n",
       "LineNumber                    0\n",
       "Quantity                      0\n",
       "Amount                        0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Index(['CustomerKey', 'EnglishProductCategoryName', 'Model', 'Country',\n",
       "       'Region', 'Age', 'IncomeGroup', 'CalendarYear', 'OrderDate',\n",
       "       'OrderNumber', 'LineNumber', 'Quantity', 'Amount'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "CustomerKey                            int64\n",
       "EnglishProductCategoryName            object\n",
       "Model                                 object\n",
       "Country                               object\n",
       "Region                                object\n",
       "Age                                    int64\n",
       "IncomeGroup                           object\n",
       "CalendarYear                           int64\n",
       "OrderDate                     datetime64[ns]\n",
       "OrderNumber                           object\n",
       "LineNumber                             int64\n",
       "Quantity                               int64\n",
       "Amount                               float64\n",
       "dtype: object"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aw_df.isnull().sum()\n",
    "aw_df.columns\n",
    "aw_df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "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>EnglishProductCategoryName</th>\n",
       "      <th>Model</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>54771</td>\n",
       "      <td>54771</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>3</td>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>Accessories</td>\n",
       "      <td>Sport-100</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>35986</td>\n",
       "      <td>6414</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       EnglishProductCategoryName                           Model\n",
       "count                       54771                           54771\n",
       "unique                          3                              37\n",
       "top          Accessories           Sport-100                     \n",
       "freq                        35986                            6414"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Series([], dtype: float64)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "EnglishProductCategoryName    False\n",
       "Model                         False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>Country</th>\n",
       "      <th>Region</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>54771</td>\n",
       "      <td>54771</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>unique</th>\n",
       "      <td>6</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>top</th>\n",
       "      <td>United States</td>\n",
       "      <td>North America</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>freq</th>\n",
       "      <td>19675</td>\n",
       "      <td>26952</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Country           Region\n",
       "count             54771            54771\n",
       "unique                6                3\n",
       "top     United States    North America  \n",
       "freq              19675            26952"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Series([], dtype: float64)"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Country    False\n",
       "Region     False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "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>Age</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>54771.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>mean</th>\n",
       "      <td>51.304157</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>std</th>\n",
       "      <td>11.437086</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>min</th>\n",
       "      <td>34.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25%</th>\n",
       "      <td>42.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>50%</th>\n",
       "      <td>49.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75%</th>\n",
       "      <td>59.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>max</th>\n",
       "      <td>104.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Age\n",
       "count  54771.000000\n",
       "mean      51.304157\n",
       "std       11.437086\n",
       "min       34.000000\n",
       "25%       42.000000\n",
       "50%       49.000000\n",
       "75%       59.000000\n",
       "max      104.000000"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Age    17.0\n",
       "dtype: float64"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Age             True\n",
       "IncomeGroup    False\n",
       "dtype: bool"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#dealing with outliers\n",
    "\n",
    "product_df = ['EnglishProductCategoryName', 'Model']\n",
    "product_df_num = aw_df[product_df]\n",
    "product_df_num.describe()\n",
    "\n",
    "Q1 =product_df_num.quantile(0.25)\n",
    "Q3 = product_df_num.quantile(0.75)\n",
    "IQR = Q3 - Q1\n",
    "IQR\n",
    "((product_df_num < (Q1 - 1.5 * IQR)) |(product_df_num > (Q3 + 1.5 * IQR))).any()\n",
    "\n",
    "\n",
    "region_df = ['Country','Region']\n",
    "region_df_num = aw_df[region_df]\n",
    "region_df_num.describe()\n",
    "\n",
    "Q1 =region_df_num.quantile(0.25)\n",
    "Q3 = region_df_num.quantile(0.75)\n",
    "IQR = Q3 - Q1\n",
    "IQR\n",
    "((region_df_num < (Q1 - 1.5 * IQR)) |(region_df_num > (Q3 + 1.5 * IQR))).any()\n",
    "\n",
    "\n",
    "\n",
    "demographic_df = ['Age', 'IncomeGroup']\n",
    "demographic_df_num = aw_df[demographic_df]\n",
    "demographic_df_num.describe()\n",
    "\n",
    "Q1 =demographic_df_num.quantile(0.25)\n",
    "Q3 = demographic_df_num.quantile(0.75)\n",
    "IQR = Q3 - Q1\n",
    "IQR\n",
    "((demographic_df_num < (Q1 - 1.5 * IQR)) |(demographic_df_num > (Q3 + 1.5 * IQR))).any()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "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>CustomerKey</th>\n",
       "      <th>EnglishProductCategoryName</th>\n",
       "      <th>Model</th>\n",
       "      <th>Country</th>\n",
       "      <th>Region</th>\n",
       "      <th>Age</th>\n",
       "      <th>IncomeGroup</th>\n",
       "      <th>CalendarYear</th>\n",
       "      <th>OrderDate</th>\n",
       "      <th>OrderNumber</th>\n",
       "      <th>LineNumber</th>\n",
       "      <th>Quantity</th>\n",
       "      <th>Amount</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11000</td>\n",
       "      <td>Bikes</td>\n",
       "      <td>Mountain-200</td>\n",
       "      <td>Australia</td>\n",
       "      <td>Pacific</td>\n",
       "      <td>49</td>\n",
       "      <td>High</td>\n",
       "      <td>2013</td>\n",
       "      <td>2013-01-18</td>\n",
       "      <td>SO51522</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2319.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11000</td>\n",
       "      <td>Accessories</td>\n",
       "      <td>Fender Set - Mountain</td>\n",
       "      <td>Australia</td>\n",
       "      <td>Pacific</td>\n",
       "      <td>49</td>\n",
       "      <td>High</td>\n",
       "      <td>2013</td>\n",
       "      <td>2013-01-18</td>\n",
       "      <td>SO51522</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>21.98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11000</td>\n",
       "      <td>Bikes</td>\n",
       "      <td>Touring-1000</td>\n",
       "      <td>Australia</td>\n",
       "      <td>Pacific</td>\n",
       "      <td>49</td>\n",
       "      <td>High</td>\n",
       "      <td>2013</td>\n",
       "      <td>2013-05-03</td>\n",
       "      <td>SO57418</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>2384.07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>11000</td>\n",
       "      <td>Accessories</td>\n",
       "      <td>Touring Tire</td>\n",
       "      <td>Australia</td>\n",
       "      <td>Pacific</td>\n",
       "      <td>49</td>\n",
       "      <td>High</td>\n",
       "      <td>2013</td>\n",
       "      <td>2013-05-03</td>\n",
       "      <td>SO57418</td>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>28.99</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11000</td>\n",
       "      <td>Accessories</td>\n",
       "      <td>Touring Tire Tube</td>\n",
       "      <td>Australia</td>\n",
       "      <td>Pacific</td>\n",
       "      <td>49</td>\n",
       "      <td>High</td>\n",
       "      <td>2013</td>\n",
       "      <td>2013-05-03</td>\n",
       "      <td>SO57418</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>4.99</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CustomerKey EnglishProductCategoryName                           Model  \\\n",
       "0        11000       Bikes                 Mountain-200                     \n",
       "1        11000       Accessories           Fender Set - Mountain            \n",
       "2        11000       Bikes                 Touring-1000                     \n",
       "3        11000       Accessories           Touring Tire                     \n",
       "4        11000       Accessories           Touring Tire Tube                \n",
       "\n",
       "           Country           Region  Age IncomeGroup  CalendarYear  OrderDate  \\\n",
       "0  Australia        Pacific           49    High              2013 2013-01-18   \n",
       "1  Australia        Pacific           49    High              2013 2013-01-18   \n",
       "2  Australia        Pacific           49    High              2013 2013-05-03   \n",
       "3  Australia        Pacific           49    High              2013 2013-05-03   \n",
       "4  Australia        Pacific           49    High              2013 2013-05-03   \n",
       "\n",
       "  OrderNumber  LineNumber  Quantity   Amount  \n",
       "0  SO51522              1         1  2319.99  \n",
       "1  SO51522              2         1    21.98  \n",
       "2  SO57418              1         1  2384.07  \n",
       "3  SO57418              2         1    28.99  \n",
       "4  SO57418              3         1     4.99  "
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "reference_date = pd.to_datetime('2013-12-01')   # reference date\n",
    "\n",
    "aw_df1 = aw_df.loc[(aw_df['OrderDate'] < '2013-12-01')]\n",
    "aw_df1.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['CustomerKey', 'EnglishProductCategoryName', 'Model', 'Country',\n",
       "       'Region', 'Age', 'IncomeGroup', 'CalendarYear', 'OrderDate',\n",
       "       'OrderNumber', 'LineNumber', 'Quantity', 'Amount'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aw_df1.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1cc209df7c0>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1cc1dafd730>"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAA2oAAAGVCAYAAACPcALuAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nOzde7hddX3v+/eHJAJyk0v0AIEm0qByM0ikbNCK0o0B60EtaChyE82GQq3dthbb/RR6Ws7R7QWkVmkU5FIEKUKhihe2gHhBINDIXUwBIYVCRKtQBU34nj/mSDqzWFlrBtaacyTz/Xqe+cwxf+P3G/M71iLry3dcfiNVhSRJkiSpPTYYdACSJEmSpNVZqEmSJElSy1ioSZIkSVLLWKhJkiRJUstYqEmSJElSy1ioSZIkSVLLDLRQS3JOkseS3NFD39OTLG5e9yb5j37EKEmSJEn9lkE+Ry3JbwNPAudX1W5rMe4PgT2r6l2TFpwkSZIkDchAz6hV1fXAT7rbkuyU5KtJbknyrSQvH2Xo4cBFfQlSkiRJkvps6qADGMVC4Piq+mGS3wI+Bbxh5cokvwHMAq4ZUHySJEmSNKlaVagl2RTYF/jHJCubNxzRbT5waVWt6GdskiRJktQvrSrU6FyK+R9VNWeMPvOBE/sUjyRJkiT1Xaum56+qnwP3JzkMIB2vXLk+ycuALYEbBhSiJEmSJE26QU/PfxGdoutlSZYmOQ44AjguyfeBO4FDuoYcDlxcg5yqUpIkSZIm2UCn55ckSZIkPdu4Z9R6eSh1kv2bB1HfmeSbExuiJEmSJA2Xcc+ojfdQ6iQvAr4LzKuqB5O8uKoem5RoJUmSJGkIjDvrY1Vdn2TmGF1+H7isqh5s+vdUpG2zzTY1c+ZYm5UkrS9uueWWH1fV9EHHsa4wR0rScBgrP07E9Pw7A9OSXAdsBnyiqs4fb9DMmTNZtGjRBHy9JKntkvxo0DGsS8yRkjQcxsqPE1GoTQX2Ag4ANgZuSPK9qrp3lEAWAAsAdtxxxwn4akmSJEla/0zE9PxLga9W1X9W1Y+B64FXjtaxqhZW1dyqmjt9ulfASJIkSdJoJqJQuwJ4bZKpSV4I/BZw9wRsV5IkSZKG0riXPjYPpd4f2CbJUuAUYBpAVZ1VVXcn+SpwG/AM8NmqWuNU/pK0Lvr1r3/N0qVLeeqppwYdSqtttNFGzJgxg2nTpg06FEnSJDAfPjfPJT/2Muvj4T30+QjwkZ6/VZLWMUuXLmWzzTZj5syZJBl0OK1UVTz++OMsXbqUWbNmDTocSdIkMB+uveeaHyfi0kdJWu899dRTbL311ialMSRh66239iirJK3HzIdr77nmRws1SeqRSWl8/owkaf3n3/q191x+ZhZqkrQOufzyy0nCPffcM7AYzjjjDH7xi18M7PslSVq6dCmHHHIIs2fP5qUvfSknnXQSTz/99IR+x3XXXcd3v/vdVZ/POusszj+/87joc889l4cffnhCv2+kiXiOmiQNnZknf3lCt/fAh97UU7+LLrqI17zmNVx88cWceuqpExpDr8444wze+c538sIXvnAg3y9Jao9B5MOq4m1vexsnnHACV1xxBStWrGDBggV84AMf4BOf+MSExXLdddex6aabsu+++wJw/PHHr1p37rnnsttuu7HddttN2PeN5Bk1SVpHPPnkk3znO9/h7LPP5uKLLwY6SeR1r3sdb3/729l55505+eSTufDCC9l7773Zfffd+dd//VcAfvSjH3HAAQewxx57cMABB/Dggw8CcMwxx3DppZeu+o5NN9101Xb3339/Dj30UF7+8pdzxBFHUFWceeaZPPzww7z+9a/n9a9/fZ9/ApIkwTXXXMNGG23EscceC8CUKVM4/fTTOf/88/nkJz/JSSedtKrv7/7u73LdddcBcMIJJzB37lx23XVXTjnllFV9Zs6cySmnnMKrXvUqdt99d+655x4eeOABzjrrLE4//XTmzJnDt771LU499VQ++tGPcumll7Jo0SKOOOII5syZw5e//GXe+ta3rtre1Vdfzdve9rbnvZ8WapK0jvinf/on5s2bx84778xWW23FrbfeCsD3v/99PvGJT3D77bdzwQUXcO+993LTTTfx7ne/m7/9278F4KSTTuKoo47itttu44gjjuC9733vuN/3L//yL5xxxhncdddd3HfffXznO9/hve99L9tttx3XXnst11577aTuryRJo7nzzjvZa6+9VmvbfPPNmTlzJsuXL1/juNNOO41FixZx22238c1vfpPbbrtt1bptttmGW2+9lRNOOIGPfvSjzJw5k+OPP54//uM/ZvHixbz2ta9d1ffQQw9l7ty5XHjhhSxevJiDDz6Yu+++m2XLlgHwuc99blUR+Xx46aOGzkSfol+X9Hp5ndrpoosu4n3vex8A8+fP56KLLuJNb3oTr371q9l2220B2GmnnTjwwAMB2H333VcVUzfccAOXXXYZAEceeSQf+MAHxv2+vffemxkzZgAwZ84cHnjgAV7zmtdM+H6pXYb5byT4d1JaF1TVqJNzVNWY4y655BIWLlzI8uXLeeSRR7jrrrvYY489AFadAdtrr71W5cteJeHII4/kH/7hHzj22GO54YYbVt3L9nxYqEnSOuDxxx/nmmuu4Y477iAJK1asIAkHH3wwG2644ap+G2ywwarPG2ywwRqPLK5McFOnTuWZZ54BOgnuV7/61ao+3dudMmXKmEcpJUnql1133ZUvfvGLq7X9/Oc/59FHH2Xrrbfm3nvvXdW+ckr8+++/n49+9KPcfPPNbLnllhxzzDGrTZe/Muc913x37LHH8uY3v5mNNtqIww47jKlTn3+Z5aWPkrQOuPTSSznqqKP40Y9+xAMPPMBDDz3ErFmz+Pa3v93T+H333XfVfW0XXnjhqjNjM2fO5JZbbgHgiiuu4Ne//vW429pss8144oknnuOeSJL0/BxwwAH84he/WHXWasWKFbz//e/npJNOYtasWSxevJhnnnmGhx56iJtuugnoFHKbbLIJW2yxBY8++ihf+cpXxv2esfLdyHXbbbcd2223HX/zN3/DMccc8/x3Egs1SVonXHTRRavdqAzwe7/3e3z+85/vafyZZ57J5z73OfbYYw8uuOCCVbNivec97+Gb3/wme++9NzfeeCObbLLJuNtasGABBx100HoxmUiSjZLclOT7Se5M8ldN+6lJ/i3J4uZ1cNeYDyZZkuQHSd7Y1b5XktubdWemOW2ZZMMkX2jab0wys9/7KUnrkyRcfvnlXHrppcyePZutt96aDTbYgL/4i79gv/32Y9asWey+++78yZ/8Ca961asAeOUrX8mee+7Jrrvuyrve9S7222+/cb/nzW9+M5dffvmqyUS6HXPMMRx//PHMmTOHX/7ylwAcccQR7LDDDuyyyy4Ts5/jXcs5WebOnVuLFi0ayHdruA3z/Rfee/Hc3X333bziFa8YdBjrhNF+Vkluqaq5AwppjZpiapOqejLJNODbwB8B84Anq+qjI/rvAlwE7A1sB/wfYOeqWpHkpmbs94CrgDOr6itJ/gDYo6qOTzIfeGtVvWOsuAaZI4f5byT4d1IaTxvz4Xe/+10OP/xwLrvssmdNMtJPJ510EnvuuSfHHXfcqOvXNj96j5okaWhV52jlk83Hac1rrCOYhwAXV9XTwP1JlgB7J3kA2LyqbgBIcj7wFuArzZhTm/GXAp9MkhrUkVJJWs/su+++/OhHPxpoDHvttRebbLIJH/vYxyZsm176KEkaakmmJFkMPAZcXVU3NqtOSnJbknOSbNm0bQ881DV8adO2fbM8sn21MVW1HPgZsPWk7IwkaSBuueUWrr/++tUm4nq+LNQkSUOtqlZU1RxgBp2zY7sBnwZ2AuYAjwArD5E+ez7ozhm4NbWPNWY1SRYkWZRk0cpn8UiShpeFmiT1yCvVxrcu/4yq6j+A64B5VfVoU8A9A3yGzj1p0DlTtkPXsBnAw037jFHaVxuTZCqwBfCTUb5/YVXNraq506dPn7D9kqSJti7/rR+U5/Izs1CTpB5stNFGPP744yanMVQVjz/+OBtttNGgQ+lZkulJXtQsbwz8DnBPkm27ur0VuKNZvhKY38zkOAuYDdxUVY8ATyTZp5mg5Cjgiq4xRzfLhwLXeH+apHWV+XDtPdf86GQiktSDGTNmsHTpUrwkbWwbbbQRM2bMGL9je2wLnJdkCp2Dl5dU1ZeSXJBkDp1LFB8A/gdAVd2Z5BLgLmA5cGJVrWi2dQJwLrAxnUlEVj6k52zggmbikZ8A8/uxY5I0GcyHz81zyY8WapLUg2nTpjFr1qxBh6EJVlW3AXuO0n7kGGNOA04bpX0RsNso7U8Bhz2/SCWpHcyH/eOlj5IkSZLUMhZqkiRJktQyFmqSJEmS1DIWapIkSZLUMhZqkiRJktQyFmqSJEmS1DIWapIkSZLUMuMWaknOSfJYkjvG6ffqJCuSHDpx4UmSJEnS8OnljNq5wLyxOiSZAnwY+NoExCRJkiRJQ23cQq2qrgd+Mk63PwS+CDw2EUFJkiRJ0jB73veoJdkeeCtw1vMPR5IkSZI0EZOJnAH8WVWtGK9jkgVJFiVZtGzZsgn4akmSJEla/0ydgG3MBS5OArANcHCS5VX1TyM7VtVCYCHA3LlzawK+W5IkSZLWO8+7UKuqWSuXk5wLfGm0Ik2SJEmS1JtxC7UkFwH7A9skWQqcAkwDqCrvS5MkSZKkCTZuoVZVh/e6sao65nlFI0mSJEmakMlEJEmSJEkTyEJNkiRJklrGQk2SJEmSWsZCTZIkSZJaxkJNkiRJklrGQk2SJEmSWsZCTZIkSZJaxkJNkiRJklrGQk2SJEmSWsZCTZIkSZJaxkJNkiRJklrGQk2SNLSSbJTkpiTfT3Jnkr9q2rdKcnWSHzbvW3aN+WCSJUl+kOSNXe17Jbm9WXdmkjTtGyb5QtN+Y5KZ/d5PSdK6x0JNkjTMngbeUFWvBOYA85LsA5wMfKOqZgPfaD6TZBdgPrArMA/4VJIpzbY+DSwAZjeveU37ccBPq+o3gdOBD/djxyRJ6zYLNUnS0KqOJ5uP05pXAYcA5zXt5wFvaZYPAS6uqqer6n5gCbB3km2Bzavqhqoq4PwRY1Zu61LggJVn2yRJWhMLNUnSUEsyJcli4DHg6qq6EXhJVT0C0Ly/uOm+PfBQ1/ClTdv2zfLI9tXGVNVy4GfA1pOzN5Kk9YWFmiRpqFXViqqaA8ygc3ZstzG6j3YmrMZoH2vM6htOFiRZlGTRsmXLxgtbkrSes1CTJAmoqv8ArqNzb9mjzeWMNO+PNd2WAjt0DZsBPNy0zxilfbUxSaYCWwA/GeX7F1bV3KqaO3369AnaK0nSuspCTZI0tJJMT/KiZnlj4HeAe4ArgaObbkcDVzTLVwLzm5kcZ9GZNOSm5vLIJ5Ls09x/dtSIMSu3dShwTXMfmyRJazR10AFIkjRA2wLnNTM3bgBcUlVfSnIDcEmS44AHgcMAqurOJJcAdwHLgROrakWzrROAc4GNga80L4CzgQuSLKFzJm1+X/ZMkrROs1CTJA2tqroN2HOU9seBA9Yw5jTgtFHaFwHPur+tqp6iKfQkSeqVlz5KkiRJUstYqEmSJElSy1ioSZIkSVLLWKhJkiRJUstYqEmSJElSy4xbqCU5J8ljSe5Yw/ojktzWvL6b5JUTH6YkSZIkDY9ezqidC8wbY/39wOuqag/gr4GFExCXJEmSJA2tcZ+jVlXXJ5k5xvrvdn38HjDj+YclSZIkScNrou9ROw74ygRvU5IkSZKGyrhn1HqV5PV0CrXXjNFnAbAAYMcdd5yor5YkSZKk9cqEnFFLsgfwWeCQqnp8Tf2qamFVza2qudOnT5+Ir5YkSZKk9c7zLtSS7AhcBhxZVfc+/5AkSZIkabiNe+ljkouA/YFtkiwFTgGmAVTVWcBfAlsDn0oCsLyq5k5WwJIkSZK0vutl1sfDx1n/buDdExaRJEmSJA25iZ71UZIkSZL0PFmoSZIkSVLLWKhJkiRJUstYqEmSJElSy1ioSZIkSVLLWKhJkiRJUstYqEmSJElSy1ioSZIkSVLLWKhJkiRJUstYqEmSJElSy1ioSZIkSVLLWKhJkiRJUstYqEmSJElSy1ioSZKGVpIdklyb5O4kdyb5o6b91CT/lmRx8zq4a8wHkyxJ8oMkb+xq3yvJ7c26M5Okad8wyRea9huTzOz3fkqS1j0WapKkYbYceH9VvQLYBzgxyS7NutOrak7zugqgWTcf2BWYB3wqyZSm/6eBBcDs5jWvaT8O+GlV/SZwOvDhPuyXJGkdZ6EmSRpaVfVIVd3aLD8B3A1sP8aQQ4CLq+rpqrofWALsnWRbYPOquqGqCjgfeEvXmPOa5UuBA1aebZMkaU0s1CRJAppLEvcEbmyaTkpyW5JzkmzZtG0PPNQ1bGnTtn2zPLJ9tTFVtRz4GbD1JOyCJGk9YqEmSRp6STYFvgi8r6p+Tucyxp2AOcAjwMdWdh1leI3RPtaYkTEsSLIoyaJly5at5R5IktY3FmqSpKGWZBqdIu3CqroMoKoeraoVVfUM8Blg76b7UmCHruEzgIeb9hmjtK82JslUYAvgJyPjqKqFVTW3quZOnz59onZPkrSOslCTJA2t5l6xs4G7q+rjXe3bdnV7K3BHs3wlML+ZyXEWnUlDbqqqR4AnkuzTbPMo4IquMUc3y4cC1zT3sUmStEZTBx2AJEkDtB9wJHB7ksVN258DhyeZQ+cSxQeA/wFQVXcmuQS4i86MkSdW1Ypm3AnAucDGwFeaF3QKwQuSLKFzJm3+JO+TJGk9YKEmSRpaVfVtRr+H7KoxxpwGnDZK+yJgt1HanwIOex5hSpKGkJc+SpIkSVLLWKhJkiRJUsuMW6g1z495LMkda1ifJGcmWdI8b+ZVEx+mJEmSJA2PXs6onQvMG2P9QXRmvZoNLKDz7BlJkiRJ0nM0bqFWVdczyvNeuhwCnF8d3wNeNGJaY0mSJEnSWpiIe9S2Bx7q+ry0aZMkSZIkPQcTUaiNNq3xqA/yTLIgyaIki5YtWzYBXy1JkiRJ65+JKNSWAjt0fZ4BPDxax6paWFVzq2ru9OnTJ+CrJUmSJGn9MxGF2pXAUc3sj/sAP6uqRyZgu5IkSZI0lKaO1yHJRcD+wDZJlgKnANMAquos4CrgYGAJ8Avg2MkKVpIkSZKGwbiFWlUdPs76Ak6csIgkSZIkachNxKWPkiRJkqQJZKEmSZIkSS1joSZJkiRJLWOhJkmSJEktY6EmSZIkSS1joSZJkiRJLWOhJkmSJEktY6EmSZIkSS1joSZJkiRJLWOhJkmSJEktY6EmSZIkSS1joSZJkiRJLWOhJkmSJEktY6EmSZIkSS1joSZJkiRJLWOhJkmSJEktY6EmSRpaSXZIcm2Su5PcmeSPmvatklyd5IfN+5ZdYz6YZEmSHyR5Y1f7Xklub9admSRN+4ZJvtC035hkZr/3U5K07rFQkyQNs+XA+6vqFcA+wIlJdgFOBr5RVbOBbzSfadbNB3YF5gGfSjKl2dangQXA7OY1r2k/DvhpVf0mcDrw4X7smCRp3WahJkkaWlX1SFXd2iw/AdwNbA8cApzXdDsPeEuzfAhwcVU9XVX3A0uAvZNsC2xeVTdUVQHnjxizcluXAgesPNsmSdKaWKhJkgQ0lyTuCdwIvKSqHoFOMQe8uOm2PfBQ17ClTdv2zfLI9tXGVNVy4GfA1pOxD5Kk9YeFmiRp6CXZFPgi8L6q+vlYXUdpqzHaxxozMoYFSRYlWbRs2bLxQpYkrecs1CRJQy3JNDpF2oVVdVnT/GhzOSPN+2NN+1Jgh67hM4CHm/YZo7SvNibJVGAL4Ccj46iqhVU1t6rmTp8+fSJ2TZK0DrNQkyQNreZesbOBu6vq412rrgSObpaPBq7oap/fzOQ4i86kITc1l0c+kWSfZptHjRizcluHAtc097FJkrRGUwcdgCRJA7QfcCRwe5LFTdufAx8CLklyHPAgcBhAVd2Z5BLgLjozRp5YVSuacScA5wIbA19pXtApBC9IsoTOmbT5k71TkqR1n4WaJGloVdW3Gf0eMoAD1jDmNOC0UdoXAbuN0v4UTaEnSVKverr0Mcm85sGeS5KcPMr6LZL8c5LvNw8MPXbiQ5UkSZKk4TBuodY8yPPvgIOAXYDDmwd+djsRuKuqXgnsD3wsyQsmOFZJkiRJGgq9nFHbG1hSVfdV1a+Ai+k8vLNbAZs1N1BvSuca/OUTGqkkSZIkDYleCrU1Pdyz2yeBV9CZivh24I+q6pkJiVCSJEmShkwvhVovD+p8I7AY2A6YA3wyyebP2pAP85QkSZKkcfVSqK3p4Z7djgUuq44lwP3Ay0duyId5SpIkSdL4einUbgZmJ5nVTBAyn87DO7s9SDONcZKXAC8D7pvIQCVJkiRpWIz7HLWqWp7kJOBrwBTgnOaBn8c3688C/ho4N8ntdC6V/LOq+vEkxi1JkiRJ662eHnhdVVcBV41oO6tr+WHgwIkNTZIkSZKGU08PvJYkSZIk9Y+FmiRJkiS1jIWaJEmSJLWMhZokSZIktYyFmiRJkiS1jIWaJEmSJLWMhZokSZIktYyFmiRJkiS1jIWaJEmSJLWMhZokSZIktYyFmiRJkiS1jIWaJEmSJLWMhZokSZIktYyFmiRJkiS1jIWaJEmSJLWMhZokSZIktYyFmiRJkiS1jIWaJEmSJLWMhZokaWglOSfJY0nu6Go7Ncm/JVncvA7uWvfBJEuS/CDJG7va90pye7PuzCRp2jdM8oWm/cYkM/u5f5KkdZeFmiRpmJ0LzBul/fSqmtO8rgJIsgswH9i1GfOpJFOa/p8GFgCzm9fKbR4H/LSqfhM4HfjwZO2IJGn9YqEmSRpaVXU98JMeux8CXFxVT1fV/cASYO8k2wKbV9UNVVXA+cBbusac1yxfChyw8mybJEljsVCTJOnZTkpyW3Np5JZN2/bAQ119ljZt2zfLI9tXG1NVy4GfAVtPZuCSpPWDhZokSav7NLATMAd4BPhY0z7ambAao32sMc+SZEGSRUkWLVu2bO0iliStdyzUJEnqUlWPVtWKqnoG+Aywd7NqKbBDV9cZwMNN+4xR2lcbk2QqsAVruNSyqhZW1dyqmjt9+vSJ2h1J0jrKQk2SpC7NPWcrvRVYOSPklcD8ZibHWXQmDbmpqh4BnkiyT3P/2VHAFV1jjm6WDwWuae5jkyRpTFN76ZRkHvAJYArw2ar60Ch99gfOAKYBP66q101gnJIkTbgkFwH7A9skWQqcAuyfZA6dSxQfAP4HQFXdmeQS4C5gOXBiVa1oNnUCnRkkNwa+0rwAzgYuSLKEzpm0+ZO/V5Kk9cG4hVoz9fDfAf+dziUcNye5sqru6urzIuBTwLyqejDJiycrYEmSJkpVHT5K89lj9D8NOG2U9kXAbqO0PwUc9nxilCQNp14ufdwbWFJV91XVr4CL6Uw33O33gcuq6kGAqnpsYsOUJEmSpOHRS6G2pumIu+0MbJnkuiS3JDlqogKUJEmSpGHTyz1qvUwtPBXYCziAzvX5NyT5XlXdu9qGkgXAAoAdd9xx7aOVJEmSpCHQyxm1NU1HPLLPV6vqP6vqx8D1wCtHbsiphyVJkiRpfL0UajcDs5PMSvICOjNWXTmizxXAa5NMTfJC4LeAuyc2VEmSJEkaDuNe+lhVy5OcBHyNzvT85zRTFB/frD+rqu5O8lXgNuAZOlP437HmrUqSJEmS1qSn56hV1VXAVSPazhrx+SPARyYuNEmSJEkaTr1c+ihJkiRJ6iMLNUmSJElqGQs1SZIkSWoZCzVJkiRJahkLNUmSJElqGQs1SZIkSWoZCzVJkiRJahkLNUmSJElqGQs1SZIkSWoZCzVJkiRJahkLNUmSJElqGQs1SZIkSWoZCzVJkiRJahkLNUmSJElqGQs1SZIkSWoZCzVJkiRJahkLNUmSJElqGQs1SZIkSWoZCzVJkiRJahkLNUmSJElqGQs1SdLQSnJOkseS3NHVtlWSq5P8sHnfsmvdB5MsSfKDJG/sat8rye3NujOTpGnfMMkXmvYbk8zs5/5JktZdFmqSpGF2LjBvRNvJwDeqajbwjeYzSXYB5gO7NmM+lWRKM+bTwAJgdvNauc3jgJ9W1W8CpwMfnrQ9kSStVyzUJElDq6quB34yovkQ4Lxm+TzgLV3tF1fV01V1P7AE2DvJtsDmVXVDVRVw/ogxK7d1KXDAyrNtkiSNxUJNkqTVvaSqHgFo3l/ctG8PPNTVb2nTtn2zPLJ9tTFVtRz4GbD1pEUuSVpvWKhJktSb0c6E1RjtY4159saTBUkWJVm0bNmy5xiiJGl90VOhlmRec+P0kiQnj9Hv1UlWJDl04kKUJKmvHm0uZ6R5f6xpXwrs0NVvBvBw0z5jlPbVxiSZCmzBsy+1BKCqFlbV3KqaO3369AnaFUnSumrqeB2aG6X/DvjvdBLOzUmurKq7Run3YeBrkxGoJEl9ciVwNPCh5v2KrvbPJ/k4sB2dSUNuqqoVSZ5Isg9wI3AU8LcjtnUDcChwTXMfm9Q6M0/+8qBDGKgHPvSmQYcgrWbcQg3YG1hSVfcBJLmYzs3Rd43o94fAF4FXT2iEkiRNkiQXAfsD2yRZCpxCp0C7JMlxwIPAYQBVdWeSS+jkv+XAiVW1otnUCXRmkNwY+ErzAjgbuCDJEjpn0ub3YbckSeuBXgq10W6e/q3uDkm2B94KvAELNUnSOqKqDl/DqgPW0P804LRR2hcBu43S/hRNoSdJ0tro5R61Xm6EPgP4s64ji6NvyBulJUmSJGlcvZxRW9PN093mAhc3j4bZBjg4yfKq+qfuTlW1EFgIMHfuXK/RlyRJkqRR9FKo3QzMTjIL+Dc619f/fneHqpq1cjnJucCXRhZpkiRJktpnmCeSafMkMuMWalW1PMlJdGZznAKc09xQfXyz/qxJjlGSJEmShkovZ9SoqquAq0a0jVqgVdUxzz8sSZIkSRpePT3wWpIkSZLUPxZqkiRJktQyFmqSJEmS1DIWapIkSZLUMhZqkiRJktQyFmqSJEmS1DIWapIkSZLUMhZqkiRJktQyFmqSJEmS1OBww3AAABfWSURBVDIWapIkSZLUMhZqkiRJktQyFmqSJEmS1DIWapIkSZLUMhZqkiRJktQyFmqSJEmS1DIWapIkSZLUMhZqkiRJktQyFmqSJEmS1DIWapIkSZLUMhZqkiRJktQyFmqSJEmS1DIWapIkSZLUMhZqkiSNIskDSW5PsjjJoqZtqyRXJ/lh875lV/8PJlmS5AdJ3tjVvleznSVJzkySQeyPJGndYqEmSdKavb6q5lTV3ObzycA3qmo28I3mM0l2AeYDuwLzgE8lmdKM+TSwAJjdvOb1MX5J0jqqp0ItybzmCOGSJCePsv6IJLc1r+8meeXEhypJ0sAdApzXLJ8HvKWr/eKqerqq7geWAHsn2RbYvKpuqKoCzu8aI0nSGo1bqDVHBP8OOAjYBTi8OXLY7X7gdVW1B/DXwMKJDlSSpD4r4OtJbkmyoGl7SVU9AtC8v7hp3x54qGvs0qZt+2Z5ZLskSWOa2kOfvYElVXUfQJKL6Rw5vGtlh6r6blf/7wEzJjJISZIGYL+qejjJi4Grk9wzRt/R7jurMdqfvYFOMbgAYMcdd1zbWCVJ65leLn1c01HCNTkO+MrzCUqSpEGrqoeb98eAy+kcuHy0uZyR5v2xpvtSYIeu4TOAh5v2GaO0j/Z9C6tqblXNnT59+kTuiiRpHdRLobY2RwNfT6dQ+7M1rF+QZFGSRcuWLes9SkmS+ijJJkk2W7kMHAjcAVwJHN10Oxq4olm+EpifZMMks+hMGnJTc3nkE0n2aWZ7PKprjCRJa9TLpY9rOkq4miR7AJ8FDqqqx0fbUFUtpLl/be7cuaMWe5IktcBLgMubmfSnAp+vqq8muRm4JMlxwIPAYQBVdWeSS+jcFrAcOLGqVjTbOgE4F9iYzhUnXnUiSRpXL4XazcDs5gjhv9GZfvj3uzsk2RG4DDiyqu6d8CglSeqj5r7sZ81g3ByIPGANY04DThulfRGw20THKElav41bqFXV8iQnAV8DpgDnNEcOj2/WnwX8JbA1nefGACzveuaMJEmSJGkt9HJGjaq6CrhqRNtZXcvvBt49saFJkiRJ0nDq6YHXkiRJkqT+sVCTJEmSpJaxUJMkSZKklrFQkyRJkqSWsVCTJEmSpJaxUJMkSZKklrFQkyRJkqSWsVCTJEmSpJaxUJMkSZKklrFQkyRJkqSWsVCTJEmSpJaxUJMkSZKklrFQkyRJkqSWsVCTJEmSpJaxUJMkSZKklrFQkyRJkqSWsVCTJEmSpJaZOugABmXmyV8edAgD88CH3jToECRJkiSNwTNqkiRJktQyFmqSJEmS1DIWapIkSZLUMhZqkiRJktQyFmqSJEmS1DJDO+ujpOHjbK+SJGld4Rk1SZIkSWqZngq1JPOS/CDJkiQnj7I+Sc5s1t+W5FUTH6okSeum8fKoJEkjjVuoJZkC/B1wELALcHiSXUZ0OwiY3bwWAJ+e4DglSVon9ZhHJUlaTS9n1PYGllTVfVX1K+Bi4JARfQ4Bzq+O7wEvSrLtBMcqSdK6qJc8KknSanop1LYHHur6vLRpW9s+kiQNI3OkJGmt9TLrY0Zpq+fQhyQL6FwaCfBkkh/08P3rq22AHw/ii/PhQXyrGgP7vYO/+wEb9t/9bww6gAEyR66dYf+3Msz83Q+vYf7/4jXmx14KtaXADl2fZwAPP4c+VNVCYGEP37neS7KoquYOOg71l7/34eXvfqiZI9eC/1aGl7/74eXvfnS9XPp4MzA7yawkLwDmA1eO6HMlcFQz++M+wM+q6pEJjlWSpHVRL3lUkqTVjHtGraqWJzkJ+BowBTinqu5Mcnyz/izgKuBgYAnwC+DYyQtZkqR1x5ry6IDDkiS1XC+XPlJVV9EpxrrbzupaLuDEiQ1tvTf0l7cMKX/vw8vf/RAbLY9qjfy3Mrz83Q8vf/ejSKfGkiRJkiS1RS/3qEmSJEmS+shCTZIkSZJaxkJNkiRJklrGQk2SJEmSWsZCTZIkSZJaxkJNkiRJklrGQk2SJEmSWsZCTZIkSZJaxkJNkiRJklrGQk2SJEmSWsZCTZIkSZJaxkJNkiRJklrGQk2SJEmSWsZCTZIkSZJaxkJNkiRJklrGQk2SJEmSWsZCTZIkSZJaxkJNkiRJklrGQk2SJEmSWsZCTZIkSZJaxkJNkiRJklrGQk2SJEmSWsZCTZIkSZJaxkJNkiRJklrGQk2SJEmSWsZCTZIkSZJaxkJNkiRJklrGQk2SJEmSWsZCTZIkSZJaxkJNkiRJklrGQm0AkmyZZI9BxyFJUpuYHyXpv1io9UmS65JsnmQr4PvA55J8fNBxSZI0SOZHSRqdhVr/bFFVPwfeBnyuqvYCfmfAMUmSNGjmR0kahYVa/0xNsi3wduBLgw5GkqSWMD9K0igs1Prn/wG+BvxrVd2c5KXADwcckyRJg2Z+lKRRpKoGHYMkSZIkqYtn1Pokyc5JvpHkjubzHkn+16DjkiRpkMyPkjQ6C7X++QzwQeDXAFV1GzB/oBFJkjR45kdJGoWFWv+8sKpuGtG2fCCRSJLUHuZHSRqFhVr//DjJTkABJDkUeGSwIUmSNHDmR0kahZOJ9Ekzi9VCYF/gp8D9wDur6oFBxiVpciT5Z5r/8ezyM2AR8PdV9VT/o5Lax/woDR9zZG8s1PosySbABlX1xKBjkTR5knwCmA5c1DS9A/h3YGNg86o6clCxSW1kfpSGhzmyNxZqkyzJO6vqH5L8z9HWV9XH+x2TpMmX5Pqq+u3R2pLcWVW7Dio2qQ3Mj9LwMkf2ZuqgAxgCmzTvmw00Ckn9Nj3JjlX1IECSHYFtmnW/GlxYUmuYH6XhZY7sgYXaJKuqv08yBfh5VZ0+6Hgk9c37gW8n+VcgwCzgD5rLu84baGRSC5gfpaFmjuyBlz72SZJrq+r1g45DUv8k2RB4OZ0kdI83R0vPZn6UhpM5cnwWan2S5DRgC+ALwH+ubK+qWwcWlKRJlWRfYCZdVy9U1fkDC0hqIfOjNJzMkeOzUOuTJNeO0lxV9Ya+ByNp0iW5ANgJWAysaJqrqt47uKik9jE/SsPHHNkbCzVJmgRJ7gZ2Kf/ISpK0GnNkbzYYdADDIskWST6eZFHz+liSLQYdlyZXkv+dZPMk05J8I8mPk7xz0HGpL+4A/q9BByG1nflxeJkjh5o5sgcWav1zDvAE8Pbm9XPgcwONSP1wYFX9HPhdYCmwM/Cngw1JfbINcFeSryW5cuVr0EFJLWR+HF7myOFljuyB0/P3z05V9Xtdn/8qyeKBRaN+mda8HwxcVFU/STLIeNQ/pw46AGkdYX4cXubI4XXqoANYF1io9c8vk7ymqr4NkGQ/4JcDjkmT75+T3EPnd/0HSaYDTj87BKrqm4OOQVpHmB+HlzlySJkje+NkIn2SZA6dB/htQed5ET8Bjqmq7w80ME26JFvSeaDriuZBjptV1b8POi5NjiTfrqrXJHkC6P4DGzozWm0+oNCkVjI/Djdz5HAxR64dC7U+S7I5QHNNttZzSV4I/E9gx6pakGQ28LKq+tKAQ5OkVjE/Dh9zpDQ2JxPpkyR/1CShJ4CPJ7k1yYGDjkuT7nPAr4B9m89Lgb8ZXDjqpyRTkmyXZMeVr0HHJLWN+XGomSOHmDlyfBZq/fOu5ijhgcCLgWOBDw02JPXBTlX1v4FfA1TVL+mc3td6LskfAo8CVwNfbl4eJZaezfw4vMyRQ8oc2RsnE+mflX94DgY+V1Xfj1MbDYNfJdmY5jrsJDsBTw82JPXJH9G5hOfxQQcitZz5cXiZI4eXObIHFmr9c0uSrwOzgA8m2Qx4ZsAxafKdAnwV2CHJhcB+wDEDjUj98hDws0EHIa0DzI/Dyxw5vMyRPXAykT5JsgEwB7ivqv4jyVbAjKq6bcChaRI1v+cA+zTv36Mzo9X9Aw1MkybJ/2wWdwVeRudyjlVHiKvq44OIS2or8+PwMkcOH3Pk2vEetf75b8APmiT0TuB/4ZGEYfDPwK+r6svNLFbTmzatvzZrXg/Sufb+BV1tmw4wLqmtzI/Dyxw5fMyRa8Ezan2S5DbglcAewAXA2cDbqup1Aw1MkyrJm4AP0Ln34uXA+cARVbV4oIFp0iU5rKr+cbw2adiZH4eXOXJ4mSN7Y6HWJ0lurapXJflL4N+q6uyVbYOOTZMryVvoJKLN6PzPxw8HHJL6YLR/3/6bl57N/DjczJHDyRzZGycT6Z8nknwQOBJ4bZIpwLQBx6RJkuRvaWaxamwO3Af8YRKq6r2DiUyTLclBdI4Ob5/kzK5VmwPLBxOV1GrmxyFjjhxe5si1Y6HWP+8Afp/O82L+vXmo30cGHJMmz6IRn28ZSBQahIfp/P7/b1b/vT8B/PFAIpLazfw4fMyRw8scuRa89LGPkvwGMLuq/k+SFwJTquqJQcclaeIlmUZnFrOdm6YfVNWvBxiS1FrmR2m4mCN74xm1PknyHmABsBWwE7A9cBZwwCDj0uRIcklVvT3J7ax+eQcAVbXHAMJSf+1L58b4B+gkox2SHF1V1w80KqllzI/DxxwpzJE98YxanyRZDOwN3FhVezZtt1fV7oONTJMhybZV9UhzlPhZqupH/Y5J/ZXkFuD3q+oHzeedgYuqaq/BRia1i/lx+JgjZY7sjWfU+ufpqvpVEgCSTGWUo0haP1TVI837qmSTZBvg8fLoyLCYtjIBAVTVvc2lHpJWZ34cMuZIYY7siQ+87p9vJvlzYOMk/x34R3yo43oryT5JrktyWZI9k9wB3AE8mmTeoONTXyxKcnaS/ZvXZ/CGeWk05schY44U5sieeOljnyTZADgOOJDOtbhfAz7rkaP1U5JFwJ8DWwALgYOq6ntJXk7n1P6eAw1Qky7JhsCJwGvo/Ju/HvhUVT090MCkljE/Dh9zpMyRvbFQ65MkmwBPVdWK5vMUYMOq+sVgI9NkSLK4quY0y3dX1Su61v2LSUiSOsyPw8ccKfXGe9T65xvA7wBPNp83Br5OZ9YbrX+e6Vr+5Yh1Hh1Zj61pFrOVnM1Mehbz4/AxRw4pc+TasVDrn42qamUSoqqebJ4Vo/XTK5P8nM7p/I2bZZrPGw0uLPXB7w46AGkdY34cPubI4WWOXAsWav3zn0leVVW3AiTZi2cfRdJ6oqqmDDoGDcw04CVV9Z3uxiSvBR4eTEhSq5kfh4w5cqiZI9eCsz72z/uAf0zyrSTfAr4AnDTgmCRNvDOAJ0Zp/2WzTtLqzI/S8DBHrgUnE+mj5vkQL6Nzav+eqvr1gEOSNMGS3FFVu61hnQ/xlUZhfpSGgzly7XhGrU+SnAhsUlV3VNXtwKZJ/mDQcUmacGPdX7Fx36KQ1hHmR2momCPXgoVa/7ynqv5j5Yeq+inwngHGI2ly3JzkWf+2kxyHD/OURmN+lIaHOXItOJlI/2yQJCsf4Nk8J+YFA45J0sR7H3B5kiP4r6Qzl86/97cOLCqpvcyP0vAwR64F71HrkyQfAWYCZ9F5fsTxwENV9f5BxiVpciR5PbDyOvw7q+qaQcYjtZX5URo+5sjeWKj1SZINgAV0HuoZOg/z/ExVPTPmQEmS1mPmR0kanYXagCR5DXB4VZ046FgkSWoL86MkdXiPWh8lmQMcDrwDuB+4bLARSZI0eOZHSXo2C7VJlmRnYD6dBPQ4nQd5pqpeP9DAJEkaIPOjJI3NSx8nWZJngG8Bx1XVkqbtvqp66WAjkyRpcMyPkjQ2n6M2+X4P+Hfg2iSfSXIAnZulJUkaZuZHSRqDZ9T6JMkmwFvoXOLxBuA84PKq+vpAA5MkaYDMj5I0Ogu1AUiyFXAY8I6qesOg45EkqQ3Mj5L0XyzUJEmSJKllvEdNkiRJklrGQk2SJEmSWsZCTRMmyYoki7teJz+PbT3ZvG+X5NIx+s1Mcsca1p2b5P4mlluT/LfnEc8xST75PMZu1/V5WpIPJflhkjuS3JTkoHG28b4kL3wu3z8Rmn14JskeXW13JJk5qJgkaV1hfhxzrPlRWgMLNU2kX1bVnK7Xh57vBqvq4ao69Hls4k+rag5wMvD3I1cmmfI8tt2rY4Dtuj7/NbAtsFtV7Qa8GdhsnG28D5jURJRk6jhdlgJ/MZkxSNJ6yvw4umMwP0prZKGmSZfkgSR/1Ry1uz3Jy5v26Umubtr/PsmPkmwzYuyqI4JJdm2Ori1OcluS2U23Kc0zeO5M8vUkG48SxvXAb3bF85dJvg0cluTwJq47kny467uPTXJvkm8C+3W1n5vk0K7PT3Ytf6DZ1vebo4KHAnOBC5u4NwHeA/xhVT0NUFWPVtUlzfhPJ1nU7MtfNW3vpZPIrk1ybdN2YJIbmp/dPybZtGk/OMk9Sb6d5MwkX2rat0ryT83P7Xsrj/wlOTXJwiRfB85P8q0kc7r25ztdRwm/BOya5GWj/I6fFXfXz/r/bWJdlORVSb6W5F+THN/V70+T3NzE91cjty9J6yPzo/nR/KgxVZUvXxPyAlYAi7te72jaH6DzhxfgD4DPNsufBD7YLM8DCtim+fxk8z4TuKNZ/lvgiGb5BcDGzfrlwJym/RLgnc3yucChzfJhwI1d8XygWd4OeBCYDkwFrqHzPJ9tu9pfAHwH+OTI7Y6I9SDgu8ALm89bNe/XAXOb5T2AfxnjZ7hyzJRm3B5dMa/82WxDJ7Fu0nz+M+AvgY2Ah4BZTftFwJe6fnanNMtvABY3y6cCtwAbN5+PBs5olncGFjXLxzS/r6OA85q2O4CZPcR9QrN8OnAbnaOj04HHmvYDgYV0HnS7AZ2E99uD/u/Zly9fvibqhfnR/Gh+9PUcXp5R00QaeWnHF7rWXda830IneQC8BrgYoKq+Cvx0nO3fAPx5kj8DfqOqftm0319Vi0fZPsBHkiwGFgDHdbWvjO3VwHVVtayqlgMXAr8N/FZX+6+6+o/ld4DPVdUvmn36SQ9jRnp7kluBfwF2BXYZpc8+Tft3mn07GvgN4OXAfVV1f9Pvoq4xrwEuaOK6Btg6yRbNuiu7fpb/CPxukmnAu+gk3W6fB/ZJMmst4r6yeb+dzv8MPFFVy4CnkryITiI6sBl7a7Mfs5Gk9Yf50fxoftRaG++aW2miPN28r+C//rvL2mygqj6f5EbgTcDXkrwbuK9r2yu3331px59W1Wg3W/9nDzGs6SGDy2kuG04SOkcUV25rvAcTLgF2TLJZVT3RvaL54/4nwKur6qdJzqVzFHCkAFdX1eEjxu85xveOtp8rY/3PVQ1Vv0hyNXAI8HY6l6XQtX55ko/ROUrZa9wrfz/PsPrv6hk6/y0E+P+q6ln3SEjSEDA/dpgf/4v5UYD3qGmwvk3njx1JDgS2HKtzkpfSOSJ2Jp2jUHuM1b9HNwKvS7JNOjdOHw58s2nfP8nWzdGzw7rGPADs1SwfAkxrlr8OvCvN7FNJtmran6C5Gbo5mng2cGaSFzT9tk3yTmBzOknhZ0leQudSEUZuA/gesF+SlfcUvDDJzsA9wEvzXzNNvaNr/PXAEU3//YEfV9XP1/Az+SxwJnDzGo56nsv/394dq0YVRHEY//42IthqoWCtj5DSp0gT1MbG0k4JoqK1KIiiRUrBLk1cBbESIShBQQt9gA1iJUFlRY/F3IUUZuPGhNzo9+sulzsMU5zDzJyZ21ZHD3XPk/r9Jx7Txm18juBoksNTtiFJ/xLzI+ZHzI//PXfUtJ0OdKUGY4OqmnQF8RXgQZJZWvAf0gLuRmaBuSTfgVXgKi0IbllVDZNcAJ7RVq6WqmoR2kFiWjnJkFZyML4B6z6wmGQZeEq34lZVg+6g8cskI2AJuEgL3HeTfAVmgHngGvAuybfu+0tV9TrJCvCWthL6fF1X7wGPkgyr6mSSM7Sx29+9n6+q90nOAYMkn4Dldd9fBhaSvAG+0MpBNhqTV0k+AwsbvB8luQXc7J4n9XtTVfUkyQngRVuAZQ2YAz5O044k9Zj50fxoftTUUrXZTrS0M7og+qMrF5gB7lS7KlhblORgVa11JSe3gQ9VdWPKNo7QDjwfr6qfO9BNSdIE5sftZ37UXuSOmnbTMeBhkn3AiHYtr/7O2SSnaecCVvjNv3EmSXIKuA6cNwlJ0q4xP24/86P2HHfUJEmSJKlnvExEkiRJknrGiZokSZIk9YwTNUmSJEnqGSdqkiRJktQzTtQkSZIkqWecqEmSJElSz/wCBmTodvZKIYoAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 1080x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# product category based on amount and quantity\n",
    "\n",
    "product_df = aw_df[['EnglishProductCategoryName','Amount']]\n",
    "product_df1 = aw_df[['EnglishProductCategoryName','Quantity']]\n",
    "\n",
    "#fig, axarr = plt.subplots(2, 2)\n",
    "#product_df.groupby(\"EnglishProductCategoryName\").sum().plot(kind=\"bar\",ax=axarr[0, 0])\n",
    "#product_df1.groupby(\"EnglishProductCategoryName\").sum().plot(kind=\"bar\",ax=axarr[0, 1])\n",
    "#plt.show()\n",
    "\n",
    "fig, axarr = plt.subplots(1, 2,figsize = (15,5))\n",
    "product_df.groupby(\"EnglishProductCategoryName\").sum().plot(kind=\"bar\",ax=axarr[0])\n",
    "product_df1.groupby(\"EnglishProductCategoryName\").sum().plot(kind=\"bar\",ax=axarr[1])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1cc1d839850>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1cc204e5070>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# which region is spending more\n",
    "\n",
    "R_group = aw_df.groupby([\"Region\"]).sum()\n",
    "total_price = R_group[\"Amount\"].groupby(level=0, group_keys=False)\n",
    "\n",
    "gtp = total_price.nlargest(5)\n",
    "\n",
    "import numpy as np\n",
    "\n",
    "# which region is has more number of customers \n",
    "\n",
    "R_C_group = aw_df.groupby([\"Region\",'CustomerKey']).sum()\n",
    "total_price = R_C_group[\"Amount\"].groupby(level=0, group_keys=False)\n",
    "\n",
    "gtp1 = total_price.nlargest(5)\n",
    "\n",
    "fig, axarr = plt.subplots(1, 2,figsize = (15,5))\n",
    "gtp.plot(kind=\"bar\",ax=axarr[0])\n",
    "gtp1.plot(kind=\"bar\",ax=axarr[1])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1cc20739130>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1cc208f4be0>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# which region/country is has more number of customers \n",
    "\n",
    "fig, axarr = plt.subplots(1, 2,figsize = (15,6))\n",
    "Customer_Country = aw_df1.groupby('Country')['CustomerKey'].nunique().sort_values(ascending=False).reset_index().head(11)\n",
    "sns.barplot(data=Customer_Country,x='Country',y='CustomerKey',palette='Blues',orient=True,ax=axarr[0])\n",
    "\n",
    "Customer_Region = aw_df1.groupby('Region')['CustomerKey'].nunique().sort_values(ascending=False).reset_index().head(11)\n",
    "sns.barplot(data=Customer_Region,x='Region',y='CustomerKey',palette='Blues',orient=True,ax=axarr[1])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>CustomerKey</th>\n",
       "      <th>Recency</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11000</td>\n",
       "      <td>212</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11001</td>\n",
       "      <td>319</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11002</td>\n",
       "      <td>281</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>11003</td>\n",
       "      <td>205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11004</td>\n",
       "      <td>214</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CustomerKey  Recency\n",
       "0        11000      212\n",
       "1        11001      319\n",
       "2        11002      281\n",
       "3        11003      205\n",
       "4        11004      214"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# create recency\n",
    "#ref_date = '2013-12-01'\n",
    "df_recency = aw_df1\n",
    "df_recency = df_recency.groupby(by='CustomerKey',as_index=False)['OrderDate'].max()\n",
    "df_recency.columns = ['CustomerKey','max_date']\n",
    "\n",
    "# the diff b/w reference date and max_date is recency\n",
    "df_recency['Recency'] = df_recency['max_date'].apply(lambda row: (reference_date - row).days)\n",
    "df_recency.drop('max_date',inplace=True,axis=1)\n",
    "df_recency[['CustomerKey','Recency']].head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 576x360 with 0 Axes>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1cc20eb0e80>"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,5))\n",
    "sns.distplot(df_recency.Recency,bins=8,kde=False,rug=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CustomerKey\n",
       "11000    5\n",
       "11001    6\n",
       "11002    2\n",
       "11003    4\n",
       "11004    3\n",
       "Name: LineNumber, dtype: int64"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# create frequency\n",
    "\n",
    "df_frequency = aw_df1\n",
    "#df_frequency = df_frequency.groupby(by='CustomerKey',as_index=False)['OrderNumber'].nunique()\n",
    "\n",
    "df_frequency = df_frequency.groupby(by='CustomerKey')['LineNumber'].nunique()\n",
    "\n",
    "\n",
    "df_frequency.columns = ['CustomerKey','Frequency']\n",
    "df_frequency.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 576x360 with 0 Axes>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1cc1d8dbdf0>"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,5))\n",
    "sns.distplot(df_frequency,bins=8,kde=False,rug=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "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>CustomerKey</th>\n",
       "      <th>Monetary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11000</td>\n",
       "      <td>4849.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11001</td>\n",
       "      <td>2419.93</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11002</td>\n",
       "      <td>2419.06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>11003</td>\n",
       "      <td>4739.30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11004</td>\n",
       "      <td>4796.02</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CustomerKey  Monetary\n",
       "0        11000   4849.00\n",
       "1        11001   2419.93\n",
       "2        11002   2419.06\n",
       "3        11003   4739.30\n",
       "4        11004   4796.02"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# create Monetary\n",
    "\n",
    "df_monetary = aw_df1\n",
    "df_monetary = df_monetary.groupby(by='CustomerKey',as_index=False)['Amount'].sum()\n",
    "\n",
    "df_monetary.columns = ['CustomerKey','Monetary']\n",
    "df_monetary.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 576x360 with 0 Axes>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1cc1e729e80>"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(8,5))\n",
    "sns.distplot(df_monetary.Monetary,kde=False,rug=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "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>CustomerKey</th>\n",
       "      <th>Recency</th>\n",
       "      <th>LineNumber</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11000</td>\n",
       "      <td>212</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11001</td>\n",
       "      <td>319</td>\n",
       "      <td>6</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11002</td>\n",
       "      <td>281</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>11003</td>\n",
       "      <td>205</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11004</td>\n",
       "      <td>214</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CustomerKey  Recency  LineNumber\n",
       "0        11000      212           5\n",
       "1        11001      319           6\n",
       "2        11002      281           2\n",
       "3        11003      205           4\n",
       "4        11004      214           3"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "CustomerKey    15834\n",
       "Recency        15834\n",
       "LineNumber     15834\n",
       "dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# we have to check RFM \n",
    "\n",
    "# lets merge Recency with Frequency\n",
    "r_f = pd.merge(df_recency,df_frequency,on='CustomerKey')\n",
    "r_f.head()\n",
    "r_f.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "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>CustomerKey</th>\n",
       "      <th>Recency</th>\n",
       "      <th>LineNumber</th>\n",
       "      <th>Monetary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11000</td>\n",
       "      <td>212</td>\n",
       "      <td>5</td>\n",
       "      <td>4849.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11001</td>\n",
       "      <td>319</td>\n",
       "      <td>6</td>\n",
       "      <td>2419.93</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11002</td>\n",
       "      <td>281</td>\n",
       "      <td>2</td>\n",
       "      <td>2419.06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>11003</td>\n",
       "      <td>205</td>\n",
       "      <td>4</td>\n",
       "      <td>4739.30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11004</td>\n",
       "      <td>214</td>\n",
       "      <td>3</td>\n",
       "      <td>4796.02</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CustomerKey  Recency  LineNumber  Monetary\n",
       "0        11000      212           5   4849.00\n",
       "1        11001      319           6   2419.93\n",
       "2        11002      281           2   2419.06\n",
       "3        11003      205           4   4739.30\n",
       "4        11004      214           3   4796.02"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "CustomerKey    15834\n",
       "Recency        15834\n",
       "LineNumber     15834\n",
       "Monetary       15834\n",
       "dtype: int64"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# we have to check RFM \n",
    "\n",
    "# lets merge Recency and Frequency with Monetary\n",
    "r_f_m = r_f.merge(df_monetary,on='CustomerKey')\n",
    "r_f_m.head()\n",
    "r_f_m.count()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "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>CustomerKey</th>\n",
       "      <th>Recency</th>\n",
       "      <th>LineNumber</th>\n",
       "      <th>Monetary</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11000</td>\n",
       "      <td>212</td>\n",
       "      <td>5</td>\n",
       "      <td>4849.00</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11001</td>\n",
       "      <td>319</td>\n",
       "      <td>6</td>\n",
       "      <td>2419.93</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11002</td>\n",
       "      <td>281</td>\n",
       "      <td>2</td>\n",
       "      <td>2419.06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>11003</td>\n",
       "      <td>205</td>\n",
       "      <td>4</td>\n",
       "      <td>4739.30</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11004</td>\n",
       "      <td>214</td>\n",
       "      <td>3</td>\n",
       "      <td>4796.02</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CustomerKey  Recency  LineNumber  Monetary\n",
       "0        11000      212           5   4849.00\n",
       "1        11001      319           6   2419.93\n",
       "2        11002      281           2   2419.06\n",
       "3        11003      205           4   4739.30\n",
       "4        11004      214           3   4796.02"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#r_f_m.set_index('CustomerKey',inplace=True)\n",
    "r_f_m.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x1cc1eddf160>"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Scatter Plot for Recency and Frequency')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Recency')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Frequency')"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Recency VS Frequrency\n",
    "\n",
    "plt.scatter(r_f_m.groupby('CustomerKey')['Recency'].sum(), aw_df1.groupby('CustomerKey')['Quantity'].sum(),\n",
    "            color = 'red',\n",
    "            marker = '*', alpha = 0.3)\n",
    "\n",
    "plt.title('Scatter Plot for Recency and Frequency') \n",
    "plt.xlabel('Recency')\n",
    "plt.ylabel('Frequency')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x1cc1e5cfac0>"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Scatter Plot for Recency and Frequency')"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Recency')"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Frequency')"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Recency Vs Frequency with Monetary\n",
    "\n",
    "Monetary = aw_df1.groupby('CustomerKey')['Amount'].sum()\n",
    "plt.scatter(r_f_m.groupby('CustomerKey')['Recency'].sum(), aw_df1.groupby('CustomerKey')['Quantity'].sum(),\n",
    "            marker = '*', alpha = 0.3,c=Monetary)\n",
    "\n",
    "plt.title('Scatter Plot for Recency and Frequency')\n",
    "plt.xlabel('Recency')\n",
    "plt.ylabel('Frequency')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.collections.PathCollection at 0x1cc1e454fa0>"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Scatter Plot for Monetary and Frequency')"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, 'Monetary')"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "Text(0, 0.5, 'Frequency')"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "#market_data = aw_df.groupby('OrderDate')[['Quantity', 'Amount']].sum()\n",
    "market_data = aw_df.groupby('CustomerKey')[['Quantity', 'Amount']].sum()\n",
    "\n",
    "plt.scatter(market_data['Amount'], market_data['Quantity'],\n",
    "            color = 'red',\n",
    "            marker = '*', alpha = 0.3)\n",
    "\n",
    "plt.title('Scatter Plot for Monetary and Frequency')\n",
    "plt.xlabel('Monetary')\n",
    "plt.ylabel('Frequency')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "def R_Score(x):\n",
    "    if x['Recency'] <=60:\n",
    "        recency = 3\n",
    "    elif x['Recency'] >60 and x['Recency'] <=120:\n",
    "        recency = 2\n",
    "    else:\n",
    "        recency = 1\n",
    "    return recency\n",
    "\n",
    "r_f_m['R'] = r_f_m.apply(R_Score,axis=1)\n",
    "\n",
    "def F_Score(x):\n",
    "    if x['LineNumber'] <= 3:\n",
    "        recency = 3\n",
    "    elif x['LineNumber'] >3 and x['LineNumber'] <=6:\n",
    "        recency = 2\n",
    "    else:\n",
    "        recency = 1\n",
    "    return recency\n",
    "\n",
    "r_f_m['F'] = r_f_m.apply(F_Score,axis=1)\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    (1120.49, 10860.57]\n",
       "1    (1120.49, 10860.57]\n",
       "2    (1120.49, 10860.57]\n",
       "3    (1120.49, 10860.57]\n",
       "4    (1120.49, 10860.57]\n",
       "Name: Monetary, dtype: category\n",
       "Categories (3, interval[float64]): [(2.289, 69.97] < (69.97, 1120.49] < (1120.49, 10860.57]]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "pd.qcut(r_f_m['Monetary'],q=3).head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "M_Score = pd.qcut(r_f_m['Monetary'],q=3,labels=range(1,4))\n",
    "r_f_m = r_f_m.assign(M = M_Score.values)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>CustomerKey</th>\n",
       "      <th>Recency</th>\n",
       "      <th>LineNumber</th>\n",
       "      <th>Monetary</th>\n",
       "      <th>R</th>\n",
       "      <th>F</th>\n",
       "      <th>M</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11000</td>\n",
       "      <td>212</td>\n",
       "      <td>5</td>\n",
       "      <td>4849.00</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11001</td>\n",
       "      <td>319</td>\n",
       "      <td>6</td>\n",
       "      <td>2419.93</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11002</td>\n",
       "      <td>281</td>\n",
       "      <td>2</td>\n",
       "      <td>2419.06</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>11003</td>\n",
       "      <td>205</td>\n",
       "      <td>4</td>\n",
       "      <td>4739.30</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11004</td>\n",
       "      <td>214</td>\n",
       "      <td>3</td>\n",
       "      <td>4796.02</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CustomerKey  Recency  LineNumber  Monetary  R  F  M\n",
       "0        11000      212           5   4849.00  1  2  3\n",
       "1        11001      319           6   2419.93  1  2  3\n",
       "2        11002      281           2   2419.06  1  3  3\n",
       "3        11003      205           4   4739.30  1  2  3\n",
       "4        11004      214           3   4796.02  1  3  3"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r_f_m.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "def RFM_Score(x):\n",
    "    return str(x['R']) + str(x['F']) + str(x['M'])\n",
    "r_f_m['RFM_Score'] = r_f_m.apply(RFM_Score,axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "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>CustomerKey</th>\n",
       "      <th>Recency</th>\n",
       "      <th>LineNumber</th>\n",
       "      <th>Monetary</th>\n",
       "      <th>R</th>\n",
       "      <th>F</th>\n",
       "      <th>M</th>\n",
       "      <th>RFM_Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11000</td>\n",
       "      <td>212</td>\n",
       "      <td>5</td>\n",
       "      <td>4849.00</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11001</td>\n",
       "      <td>319</td>\n",
       "      <td>6</td>\n",
       "      <td>2419.93</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11002</td>\n",
       "      <td>281</td>\n",
       "      <td>2</td>\n",
       "      <td>2419.06</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>11003</td>\n",
       "      <td>205</td>\n",
       "      <td>4</td>\n",
       "      <td>4739.30</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11004</td>\n",
       "      <td>214</td>\n",
       "      <td>3</td>\n",
       "      <td>4796.02</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>133</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CustomerKey  Recency  LineNumber  Monetary  R  F  M RFM_Score\n",
       "0        11000      212           5   4849.00  1  2  3       123\n",
       "1        11001      319           6   2419.93  1  2  3       123\n",
       "2        11002      281           2   2419.06  1  3  3       133\n",
       "3        11003      205           4   4739.30  1  2  3       123\n",
       "4        11004      214           3   4796.02  1  3  3       133"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r_f_m.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "#The code below allows us to create a new column “segment” which represents the segment in which our customer is located.\n",
    "#We first start to identify the segments: ‘Lost Cheap customers’,’Lost Customer,’Best Customers’, and ’Almost Customers’.\n",
    "#We assign ‘others’ for others\n",
    "\n",
    "segment = [0]*len(r_f_m)\n",
    "#r_f_m['segment'] = segment\n",
    "\n",
    "best = list(r_f_m.loc[r_f_m['RFM_Score']=='333'].index)\n",
    "lost_cheap = list(r_f_m.loc[r_f_m['RFM_Score']=='111'].index)\n",
    "lost = list(r_f_m.loc[r_f_m['RFM_Score']=='133'].index)\n",
    "lost_almost = list(r_f_m.loc[r_f_m['RFM_Score']=='233'].index)\n",
    "\n",
    "for i in range(0,len(r_f_m)):\n",
    "    if r_f_m['RFM_Score'][i]=='111':\n",
    "        segment[i]='Lost Cheap Customers'\n",
    "    elif r_f_m['RFM_Score'][i]=='133':\n",
    "        segment[i]='Lost Customers'\n",
    "    elif r_f_m['RFM_Score'][i]=='233':\n",
    "        segment[i]='Almost Lost Customers'\n",
    "    elif r_f_m['RFM_Score'][i]=='333':\n",
    "        segment[i]='Best Customers'\n",
    "    else:\n",
    "        segment[i]='Others'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "best = list(r_f_m.loc[r_f_m['RFM_Score']=='333'].index)\n",
    "lost_cheap = list(r_f_m.loc[r_f_m['RFM_Score']=='111'].index)\n",
    "lost = list(r_f_m.loc[r_f_m['RFM_Score']=='133'].index)\n",
    "lost_almost = list(r_f_m.loc[r_f_m['RFM_Score']=='233'].index)"
   ]
  },
  {
   "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>CustomerKey</th>\n",
       "      <th>Recency</th>\n",
       "      <th>LineNumber</th>\n",
       "      <th>Monetary</th>\n",
       "      <th>R</th>\n",
       "      <th>F</th>\n",
       "      <th>M</th>\n",
       "      <th>RFM_Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11000</td>\n",
       "      <td>212</td>\n",
       "      <td>5</td>\n",
       "      <td>4849.00</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11001</td>\n",
       "      <td>319</td>\n",
       "      <td>6</td>\n",
       "      <td>2419.93</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11002</td>\n",
       "      <td>281</td>\n",
       "      <td>2</td>\n",
       "      <td>2419.06</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>11003</td>\n",
       "      <td>205</td>\n",
       "      <td>4</td>\n",
       "      <td>4739.30</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11004</td>\n",
       "      <td>214</td>\n",
       "      <td>3</td>\n",
       "      <td>4796.02</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>133</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CustomerKey  Recency  LineNumber  Monetary  R  F  M RFM_Score\n",
       "0        11000      212           5   4849.00  1  2  3       123\n",
       "1        11001      319           6   2419.93  1  2  3       123\n",
       "2        11002      281           2   2419.06  1  3  3       133\n",
       "3        11003      205           4   4739.30  1  2  3       123\n",
       "4        11004      214           3   4796.02  1  3  3       133"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#The customers in the ‘Loyal customers’ segment are then identified and assigned to ‘segment’ as well\n",
    "\n",
    "loyal = list(r_f_m.loc[r_f_m['F']==3].index)\n",
    "loyal2=[]\n",
    "for i in loyal:\n",
    "    if i not in best and i not in lost and i not in lost_cheap and i not in lost_almost:\n",
    "            loyal2.append(i)\n",
    "for i in r_f_m.index:\n",
    "    if i in loyal2:\n",
    "        segment[i] = 'Loyal Customers'\n",
    "\n",
    "r_f_m.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "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>CustomerKey</th>\n",
       "      <th>Recency</th>\n",
       "      <th>LineNumber</th>\n",
       "      <th>Monetary</th>\n",
       "      <th>R</th>\n",
       "      <th>F</th>\n",
       "      <th>M</th>\n",
       "      <th>RFM_Score</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11000</td>\n",
       "      <td>212</td>\n",
       "      <td>5</td>\n",
       "      <td>4849.00</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11001</td>\n",
       "      <td>319</td>\n",
       "      <td>6</td>\n",
       "      <td>2419.93</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11002</td>\n",
       "      <td>281</td>\n",
       "      <td>2</td>\n",
       "      <td>2419.06</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>133</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>11003</td>\n",
       "      <td>205</td>\n",
       "      <td>4</td>\n",
       "      <td>4739.30</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11004</td>\n",
       "      <td>214</td>\n",
       "      <td>3</td>\n",
       "      <td>4796.02</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>133</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CustomerKey  Recency  LineNumber  Monetary  R  F  M RFM_Score\n",
       "0        11000      212           5   4849.00  1  2  3       123\n",
       "1        11001      319           6   2419.93  1  2  3       123\n",
       "2        11002      281           2   2419.06  1  3  3       133\n",
       "3        11003      205           4   4739.30  1  2  3       123\n",
       "4        11004      214           3   4796.02  1  3  3       133"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#The customers in the ‘big spenders’ segment are then identified and assigned to ‘segment’ as well\n",
    "\n",
    "big = list(r_f_m.loc[r_f_m['M']==3].index)\n",
    "big2=[]\n",
    "for i in big:\n",
    "    if i not in best and i not in lost and i not in lost_cheap and i not in lost_almost:\n",
    "            big2.append(i)\n",
    "for i in r_f_m.index:\n",
    "    if i in big2:\n",
    "        segment[i] = 'Big Spenders'\n",
    "\n",
    "r_f_m.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "r_f_m['segment'] = segment"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "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>CustomerKey</th>\n",
       "      <th>Recency</th>\n",
       "      <th>LineNumber</th>\n",
       "      <th>Monetary</th>\n",
       "      <th>R</th>\n",
       "      <th>F</th>\n",
       "      <th>M</th>\n",
       "      <th>RFM_Score</th>\n",
       "      <th>segment</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11000</td>\n",
       "      <td>212</td>\n",
       "      <td>5</td>\n",
       "      <td>4849.00</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "      <td>Big Spenders</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11001</td>\n",
       "      <td>319</td>\n",
       "      <td>6</td>\n",
       "      <td>2419.93</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "      <td>Big Spenders</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11002</td>\n",
       "      <td>281</td>\n",
       "      <td>2</td>\n",
       "      <td>2419.06</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>133</td>\n",
       "      <td>Lost Customers</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>11003</td>\n",
       "      <td>205</td>\n",
       "      <td>4</td>\n",
       "      <td>4739.30</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "      <td>Big Spenders</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11004</td>\n",
       "      <td>214</td>\n",
       "      <td>3</td>\n",
       "      <td>4796.02</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>133</td>\n",
       "      <td>Lost Customers</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>11005</td>\n",
       "      <td>213</td>\n",
       "      <td>4</td>\n",
       "      <td>4746.34</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "      <td>Big Spenders</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>11006</td>\n",
       "      <td>201</td>\n",
       "      <td>1</td>\n",
       "      <td>2384.07</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>133</td>\n",
       "      <td>Lost Customers</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>11007</td>\n",
       "      <td>257</td>\n",
       "      <td>5</td>\n",
       "      <td>4811.01</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "      <td>Big Spenders</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>11008</td>\n",
       "      <td>274</td>\n",
       "      <td>4</td>\n",
       "      <td>4731.32</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "      <td>Big Spenders</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>11009</td>\n",
       "      <td>206</td>\n",
       "      <td>2</td>\n",
       "      <td>4716.34</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>133</td>\n",
       "      <td>Lost Customers</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>11010</td>\n",
       "      <td>192</td>\n",
       "      <td>2</td>\n",
       "      <td>4688.05</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>133</td>\n",
       "      <td>Lost Customers</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>11011</td>\n",
       "      <td>257</td>\n",
       "      <td>2</td>\n",
       "      <td>4733.05</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>133</td>\n",
       "      <td>Lost Customers</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>11012</td>\n",
       "      <td>47</td>\n",
       "      <td>3</td>\n",
       "      <td>81.26</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>332</td>\n",
       "      <td>Loyal Customers</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>11013</td>\n",
       "      <td>232</td>\n",
       "      <td>2</td>\n",
       "      <td>38.98</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>131</td>\n",
       "      <td>Loyal Customers</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>11014</td>\n",
       "      <td>215</td>\n",
       "      <td>4</td>\n",
       "      <td>138.45</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>2</td>\n",
       "      <td>122</td>\n",
       "      <td>Others</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>11015</td>\n",
       "      <td>317</td>\n",
       "      <td>3</td>\n",
       "      <td>2500.97</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>133</td>\n",
       "      <td>Lost Customers</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>11016</td>\n",
       "      <td>295</td>\n",
       "      <td>3</td>\n",
       "      <td>2332.28</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>133</td>\n",
       "      <td>Lost Customers</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>11017</td>\n",
       "      <td>48</td>\n",
       "      <td>2</td>\n",
       "      <td>3059.32</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>333</td>\n",
       "      <td>Best Customers</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>11018</td>\n",
       "      <td>38</td>\n",
       "      <td>4</td>\n",
       "      <td>3133.29</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>323</td>\n",
       "      <td>Big Spenders</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>11019</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>574.25</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>2</td>\n",
       "      <td>332</td>\n",
       "      <td>Loyal Customers</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    CustomerKey  Recency  LineNumber  Monetary  R  F  M RFM_Score  \\\n",
       "0         11000      212           5   4849.00  1  2  3       123   \n",
       "1         11001      319           6   2419.93  1  2  3       123   \n",
       "2         11002      281           2   2419.06  1  3  3       133   \n",
       "3         11003      205           4   4739.30  1  2  3       123   \n",
       "4         11004      214           3   4796.02  1  3  3       133   \n",
       "5         11005      213           4   4746.34  1  2  3       123   \n",
       "6         11006      201           1   2384.07  1  3  3       133   \n",
       "7         11007      257           5   4811.01  1  2  3       123   \n",
       "8         11008      274           4   4731.32  1  2  3       123   \n",
       "9         11009      206           2   4716.34  1  3  3       133   \n",
       "10        11010      192           2   4688.05  1  3  3       133   \n",
       "11        11011      257           2   4733.05  1  3  3       133   \n",
       "12        11012       47           3     81.26  3  3  2       332   \n",
       "13        11013      232           2     38.98  1  3  1       131   \n",
       "14        11014      215           4    138.45  1  2  2       122   \n",
       "15        11015      317           3   2500.97  1  3  3       133   \n",
       "16        11016      295           3   2332.28  1  3  3       133   \n",
       "17        11017       48           2   3059.32  3  3  3       333   \n",
       "18        11018       38           4   3133.29  3  2  3       323   \n",
       "19        11019        1           3    574.25  3  3  2       332   \n",
       "\n",
       "            segment  \n",
       "0      Big Spenders  \n",
       "1      Big Spenders  \n",
       "2    Lost Customers  \n",
       "3      Big Spenders  \n",
       "4    Lost Customers  \n",
       "5      Big Spenders  \n",
       "6    Lost Customers  \n",
       "7      Big Spenders  \n",
       "8      Big Spenders  \n",
       "9    Lost Customers  \n",
       "10   Lost Customers  \n",
       "11   Lost Customers  \n",
       "12  Loyal Customers  \n",
       "13  Loyal Customers  \n",
       "14           Others  \n",
       "15   Lost Customers  \n",
       "16   Lost Customers  \n",
       "17   Best Customers  \n",
       "18     Big Spenders  \n",
       "19  Loyal Customers  "
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "r_f_m.head(20)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<Figure size 720x576 with 0 Axes>"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1cc1e27f5e0>"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sq1 = r_f_m.groupby('segment')['CustomerKey'].nunique().sort_values(ascending=False).reset_index()\n",
    "plt.figure(figsize=(10,8))\n",
    "sq1.drop([0],inplace=True)\n",
    "sns.barplot(data=sq1,x='segment',y='CustomerKey',palette='Greens_d',orient=True)"
   ]
  },
  {
   "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>segment</th>\n",
       "      <th>CustomerKey</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Lost Customers</td>\n",
       "      <td>1773</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Big Spenders</td>\n",
       "      <td>1659</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Others</td>\n",
       "      <td>1624</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Best Customers</td>\n",
       "      <td>996</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Almost Lost Customers</td>\n",
       "      <td>737</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                 segment  CustomerKey\n",
       "1         Lost Customers         1773\n",
       "2           Big Spenders         1659\n",
       "3                 Others         1624\n",
       "4         Best Customers          996\n",
       "5  Almost Lost Customers          737"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "737"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sq1\n",
    "mini = min(sq1['CustomerKey'])\n",
    "mini"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [],
   "source": [
    "import squarify\n",
    "import matplotlib\n",
    "cmap = matplotlib.cm.coolwarm\n",
    "mini = min(sq1['CustomerKey'])\n",
    "maxi = max(sq1['CustomerKey'])\n",
    "norm = matplotlib.colors.Normalize(vmin=mini, vmax=maxi)\n",
    "colors = [cmap(norm(value)) for value in sq1['CustomerKey']]\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1cc1dd68730>"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "(0.0, 100.0, 0.0, 100.0)"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1008x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "fig = plt.gcf()\n",
    "ax = fig.add_subplot()\n",
    "fig.set_size_inches(14,10)\n",
    "squarify.plot(sizes=sq1['CustomerKey'], label=sq1.segment,alpha=1,color=colors)\n",
    "plt.axis('off')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [],
   "source": [
    "r_f_m.to_excel('RFM_for Tableau.xlsx',sheet_name='RFM')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 110,
   "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>CustomerKey</th>\n",
       "      <th>Response</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11000</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11001</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11002</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>11003</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11004</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CustomerKey  Response\n",
       "0        11000         0\n",
       "1        11001         1\n",
       "2        11002         0\n",
       "3        11003         0\n",
       "4        11004         0"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "text/plain": [
       "CustomerKey    int64\n",
       "Response       int64\n",
       "dtype: object"
      ]
     },
     "execution_count": 110,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "response_df = pd.read_excel('RFM_Response.xlsx')\n",
    "response_df.head()\n",
    "response_df.dtypes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 111,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_RFM = pd.DataFrame(r_f_m)\n",
    "\n",
    "df_RFM1 = df_RFM.drop(['segment'],axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 112,
   "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>CustomerKey</th>\n",
       "      <th>Recency</th>\n",
       "      <th>LineNumber</th>\n",
       "      <th>Monetary</th>\n",
       "      <th>R</th>\n",
       "      <th>F</th>\n",
       "      <th>M</th>\n",
       "      <th>RFM_Score</th>\n",
       "      <th>segment</th>\n",
       "      <th>Response</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>11000</td>\n",
       "      <td>212</td>\n",
       "      <td>5</td>\n",
       "      <td>4849.00</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "      <td>Big Spenders</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11001</td>\n",
       "      <td>319</td>\n",
       "      <td>6</td>\n",
       "      <td>2419.93</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "      <td>Big Spenders</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>11002</td>\n",
       "      <td>281</td>\n",
       "      <td>2</td>\n",
       "      <td>2419.06</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>133</td>\n",
       "      <td>Lost Customers</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>11003</td>\n",
       "      <td>205</td>\n",
       "      <td>4</td>\n",
       "      <td>4739.30</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "      <td>Big Spenders</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>11004</td>\n",
       "      <td>214</td>\n",
       "      <td>3</td>\n",
       "      <td>4796.02</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>133</td>\n",
       "      <td>Lost Customers</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   CustomerKey  Recency  LineNumber  Monetary  R  F  M RFM_Score  \\\n",
       "0        11000      212           5   4849.00  1  2  3       123   \n",
       "1        11001      319           6   2419.93  1  2  3       123   \n",
       "2        11002      281           2   2419.06  1  3  3       133   \n",
       "3        11003      205           4   4739.30  1  2  3       123   \n",
       "4        11004      214           3   4796.02  1  3  3       133   \n",
       "\n",
       "          segment  Response  \n",
       "0    Big Spenders         0  \n",
       "1    Big Spenders         1  \n",
       "2  Lost Customers         0  \n",
       "3    Big Spenders         0  \n",
       "4  Lost Customers         0  "
      ]
     },
     "execution_count": 112,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_RFM = pd.DataFrame(r_f_m)\n",
    "\n",
    "df_RFMRv = pd.merge(df_RFM,response_df,on='CustomerKey')\n",
    "\n",
    "df_RFMRv.head()\n",
    "df_RFMRv.set_index('CustomerKey',inplace=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 113,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Label Encoding\n",
    "from sklearn import preprocessing\n",
    "label_encoder = preprocessing.LabelEncoder()\n",
    "df_RFMRv['segment']= label_encoder.fit_transform(df_RFMRv['segment']) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 114,
   "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>Recency</th>\n",
       "      <th>LineNumber</th>\n",
       "      <th>Monetary</th>\n",
       "      <th>R</th>\n",
       "      <th>F</th>\n",
       "      <th>M</th>\n",
       "      <th>RFM_Score</th>\n",
       "      <th>Response</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>CustomerKey</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>11000</th>\n",
       "      <td>212</td>\n",
       "      <td>5</td>\n",
       "      <td>4849.00</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11001</th>\n",
       "      <td>319</td>\n",
       "      <td>6</td>\n",
       "      <td>2419.93</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11002</th>\n",
       "      <td>281</td>\n",
       "      <td>2</td>\n",
       "      <td>2419.06</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>133</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11003</th>\n",
       "      <td>205</td>\n",
       "      <td>4</td>\n",
       "      <td>4739.30</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>3</td>\n",
       "      <td>123</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11004</th>\n",
       "      <td>214</td>\n",
       "      <td>3</td>\n",
       "      <td>4796.02</td>\n",
       "      <td>1</td>\n",
       "      <td>3</td>\n",
       "      <td>3</td>\n",
       "      <td>133</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             Recency  LineNumber  Monetary  R  F  M RFM_Score  Response\n",
       "CustomerKey                                                            \n",
       "11000            212           5   4849.00  1  2  3       123         0\n",
       "11001            319           6   2419.93  1  2  3       123         1\n",
       "11002            281           2   2419.06  1  3  3       133         0\n",
       "11003            205           4   4739.30  1  2  3       123         0\n",
       "11004            214           3   4796.02  1  3  3       133         0"
      ]
     },
     "execution_count": 114,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "list = ['segment']\n",
    "\n",
    "df_RFMRv.drop(df_RFMRv[list], axis=1, inplace=True)\n",
    "df_RFMRv.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Modelling "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 115,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "X_train shape: (11083, 8)\n",
      "X_test shape: (4751, 8)\n"
     ]
    }
   ],
   "source": [
    "\n",
    "from sklearn.model_selection import train_test_split\n",
    "from sklearn.linear_model import LogisticRegression\n",
    "from sklearn import metrics\n",
    "\n",
    "X_train, X_test, y_train, y_test = train_test_split(df_RFM1,df_RFMRv['Response'],test_size=0.3,random_state=0)\n",
    "\n",
    "# lets see the shape of our train and test datasets\n",
    "\n",
    "print('X_train shape: {}'.format(X_train.shape))\n",
    "print('X_test shape: {}'.format(X_test.shape))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 116,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "LogisticRegression()"
      ]
     },
     "execution_count": 116,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# now we have to get the predicted values from our trained dataset. \n",
    "# fit function will do the job\n",
    "\n",
    "logistic_regression= LogisticRegression(max_iter=100)\n",
    "logistic_regression.fit(X_train,y_train)\n",
    "y_pred=logistic_regression.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 117,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "CustomerKey    0\n",
       "Recency        0\n",
       "LineNumber     0\n",
       "Monetary       0\n",
       "R              0\n",
       "F              0\n",
       "M              0\n",
       "RFM_Score      0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 117,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#summarize missing values - X_train\n",
    "X_train.isnull().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 118,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1cc24656370>"
      ]
     },
     "execution_count": 118,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "confusion_matrix = pd.crosstab(y_test, y_pred, rownames=['Actual'], colnames=['Predicted'])\n",
    "sns.heatmap(confusion_matrix, annot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 119,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.roc_curve.RocCurveDisplay at 0x1cc1df45580>"
      ]
     },
     "execution_count": 119,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# roc plot\n",
    "from sklearn.metrics import plot_roc_curve\n",
    "fig,ax = plt.subplots(figsize=(7,7))\n",
    "plot_roc_curve(logistic_regression, X_test, y_test, ax = ax)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 120,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.95      1.00      0.98      4527\n",
      "           1       0.00      0.00      0.00       224\n",
      "\n",
      "    accuracy                           0.95      4751\n",
      "   macro avg       0.48      0.50      0.49      4751\n",
      "weighted avg       0.91      0.95      0.93      4751\n",
      "\n",
      "Accuracy:  0.9528520311513365\n",
      "0.9298471965988575\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Surendra\\anaconda3\\lib\\site-packages\\sklearn\\metrics\\_classification.py:1221: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    }
   ],
   "source": [
    "## we choose F1-score as our metric to evaluate our model.\n",
    "\n",
    "#Based on the performance metrics below, the best model is Logistic Regression with F1-score of 80%.\n",
    "\n",
    "from sklearn.metrics import classification_report, confusion_matrix\n",
    "print(classification_report(y_test,y_pred))\n",
    "\n",
    "print('Accuracy: ',metrics.accuracy_score(y_test, y_pred))\n",
    "plt.show()\n",
    "\n",
    "from sklearn.metrics import f1_score\n",
    "print(f1_score(y_test, y_pred, average = \"weighted\"))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 121,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DecisionTreeClassifier(max_depth=4, random_state=0)"
      ]
     },
     "execution_count": 121,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# training and predicting\n",
    "from sklearn.tree import DecisionTreeClassifier\n",
    "clf = DecisionTreeClassifier(random_state=0, max_depth=4)\n",
    "clf.fit(X_train,y_train)\n",
    "y_pred = clf.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 122,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.95      1.00      0.98      4527\n",
      "           1       0.70      0.03      0.06       224\n",
      "\n",
      "    accuracy                           0.95      4751\n",
      "   macro avg       0.83      0.52      0.52      4751\n",
      "weighted avg       0.94      0.95      0.93      4751\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# evaluation\n",
    "from sklearn.metrics import classification_report \n",
    "print(classification_report(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 123,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[4524    3]\n",
      " [ 217    7]]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1cc1eac7340>"
      ]
     },
     "execution_count": 123,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# confusion_matrix\n",
    "from sklearn.metrics import confusion_matrix\n",
    "print(confusion_matrix(y_test, y_pred))\n",
    "\n",
    "confusion_matrix = pd.crosstab(y_test, y_pred, rownames=['Actual'], colnames=['Predicted'])\n",
    "sns.heatmap(confusion_matrix, annot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 124,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.roc_curve.RocCurveDisplay at 0x1cc1df452b0>"
      ]
     },
     "execution_count": 124,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# roc plot\n",
    "from sklearn.metrics import plot_roc_curve\n",
    "fig,ax = plt.subplots(figsize=(7,7))\n",
    "plot_roc_curve(clf, X_test, y_test, ax = ax)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 125,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1cc258483a0>"
      ]
     },
     "execution_count": 125,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "importances = clf.feature_importances_\n",
    "weights = pd.Series(importances,index=X_train.columns.values)\n",
    "weights.sort_values()[-10:].plot(kind = 'bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 126,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RandomForestClassifier(random_state=0)"
      ]
     },
     "execution_count": 126,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Model accuracy score with 10 decision-trees : 0.9569\n"
     ]
    }
   ],
   "source": [
    "from sklearn.ensemble import RandomForestClassifier\n",
    "\n",
    "# instantiate the classifier \n",
    "\n",
    "rfc = RandomForestClassifier(random_state=0)\n",
    "\n",
    "# fit the model\n",
    "\n",
    "rfc.fit(X_train, y_train)\n",
    "\n",
    "# Predict the Test set results\n",
    "\n",
    "y_pred = rfc.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 128,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.96      1.00      0.98      4527\n",
      "           1       0.81      0.11      0.20       224\n",
      "\n",
      "    accuracy                           0.96      4751\n",
      "   macro avg       0.88      0.56      0.59      4751\n",
      "weighted avg       0.95      0.96      0.94      4751\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# evaluation\n",
    "from sklearn.metrics import classification_report \n",
    "print(classification_report(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 127,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[4521    6]\n",
      " [ 199   25]]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1cc1dfeb370>"
      ]
     },
     "execution_count": 127,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# confusion_matrix\n",
    "from sklearn.metrics import confusion_matrix\n",
    "print(confusion_matrix(y_test, y_pred))\n",
    "\n",
    "confusion_matrix = pd.crosstab(y_test, y_pred, rownames=['Actual'], colnames=['Predicted'])\n",
    "sns.heatmap(confusion_matrix, annot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 129,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.roc_curve.RocCurveDisplay at 0x1cc2579eee0>"
      ]
     },
     "execution_count": 129,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 504x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# roc plot\n",
    "from sklearn.metrics import plot_roc_curve\n",
    "fig,ax = plt.subplots(figsize=(7,7))\n",
    "plot_roc_curve(clf, X_test, y_test, ax = ax)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 130,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "GaussianNB()"
      ]
     },
     "execution_count": 130,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from sklearn.naive_bayes import GaussianNB\n",
    "\n",
    "\n",
    "# instantiate the model\n",
    "gnb = GaussianNB()\n",
    "\n",
    "# fit the model\n",
    "gnb.fit(X_train, y_train)\n",
    "\n",
    "y_pred = gnb.predict(X_test)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 131,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.95      1.00      0.98      4527\n",
      "           1       0.00      0.00      0.00       224\n",
      "\n",
      "    accuracy                           0.95      4751\n",
      "   macro avg       0.48      0.50      0.49      4751\n",
      "weighted avg       0.91      0.95      0.93      4751\n",
      "\n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\Surendra\\anaconda3\\lib\\site-packages\\sklearn\\metrics\\_classification.py:1221: UndefinedMetricWarning: Precision and F-score are ill-defined and being set to 0.0 in labels with no predicted samples. Use `zero_division` parameter to control this behavior.\n",
      "  _warn_prf(average, modifier, msg_start, len(result))\n"
     ]
    }
   ],
   "source": [
    "# evaluation\n",
    "from sklearn.metrics import classification_report \n",
    "print(classification_report(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 132,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[4527    0]\n",
      " [ 224    0]]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1cc23235d90>"
      ]
     },
     "execution_count": 132,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# confusion_matrix\n",
    "from sklearn.metrics import confusion_matrix\n",
    "print(confusion_matrix(y_test, y_pred))\n",
    "\n",
    "confusion_matrix = pd.crosstab(y_test, y_pred, rownames=['Actual'], colnames=['Predicted'])\n",
    "sns.heatmap(confusion_matrix, annot=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 133,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<sklearn.metrics._plot.roc_curve.RocCurveDisplay at 0x1cc1dd64940>"
      ]
     },
     "execution_count": 133,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbkAAAGpCAYAAAAQgkizAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deXxU9b3/8dcnO4QQlrAHCCC7AkoAl6q4IbiUal2KFZfWX69Waze91WurVVvbWq7aVq9erlorKGpdqaK4a1XCpoisAYGQsISwZ9/m+/tjhhhClknI5Exm3s/HIw/mnPOdM585kPPmnPM932POOURERCJRjNcFiIiIhIpCTkREIpZCTkREIpZCTkREIpZCTkREIlac1wU0V1pamsvIyPC6DBERCSPLly/f7ZzrUXd+uwu5jIwMli1b5nUZIiISRswsp775Ol0pIiIRSyEnIiIRSyEnIiIRSyEnIiIRSyEnIiIRSyEnIiIRSyEnIiIRSyEnIiIRSyEnIiIRSyEnIiIRSyEnIiIRSyEnIiIRK2QhZ2ZPmtkuM1vVwHIzs7+a2UYzW2lmJ4SqFhERiU6hPJJ7CpjayPJpwNDAz4+AR0NYi4iIRKGQPWrHOfexmWU00mQ68LRzzgFZZtbFzPo453aEqiYRETk6zjl8DnzOUe1zuMBrn3P4fLVeuzqvfYfPd85R7YPEuBgy0pJDVq+Xz5PrB+TWms4LzFPIiUSouju6w3Z6PqiumXf4DtM5qK71Xuecv62Pb5bXtPW3OdTe1btDJjBdtxb/OqtdnWW+2u3q+R6+Ojvvw3bm9ddy6Hv4Ajv7mte11lNfLUEHSz3zm6qlOvBdapa52nX639vaxqSnMv+mb7X6eg/xMuSsnnn1bkEz+xH+U5oMGDAglDWJtAtlldVs3FXEmh0HWbejkH0lFfXsgBvYITcQDjU7Ol/d13V2er5aO71m7oCjgRnEmBFrVvM6xiAmxr55bRaYPrTciIn55rUZxNZ6HWNGbKC9BdYRG2P+1zEQHxMTaBtYVuv1ofc2q5Za7WrXUve9VquuQ58fW/O+b2qp+z2s1md17Rgf0r8PL0MuD+hfazod2F5fQ+fcbGA2QGZmZpT8qoj4w2RXYTlrdxxk7Y7CwJ8H2bS7uOZ/1UnxMfRISWxgZ1L/juzQDjg2xoiPqbNjq3cHXHtHVmunFxPY0dXa6fk/+5ud3mE75EPvrbWjqxsG37QNZod8eEh8U0swO2QC07WW196GMXW/F0ds09h6vodZff9/F694GXLzgZvM7DlgEnBA1+MkmpVX+Y/ODoXZup3+YNtbXFHTpl+XDozoncK5o3szsk9nRvRJIaN7MrEx2rGK1CdkIWdm84DJQJqZ5QF3AfEAzrnHgAXAecBGoAS4NlS1iISjwrJK3li5g6xNe1i7o5CvC4qoChydJcbFMKJ3CueM7MXIPimM6NOZkb07kxriUzsikSaUvStnNLHcATeG6vNFwpFzjsWb9/LCslwWfLWDskofvTsnMapvZ84e1dN/dNa7M4PSdHQm0hq8PF0pEjV2HCjl5c+38cKyXHL2lJCSGMfFJ6RzWWZ/xqan6jqOSIgo5ERCpLyqmvfW7uKFZbl8nF2Az8GJg7vx07OGMu3YPnRIiPW6RJGIp5ATaWXrdh7khaV5vPJFHvtKKumTmsSNZxzDJePTGdg9dDe9isiRFHIircA5xycbd/OXdzewLGcf8bHGlFG9uTQznVOH9tD1NRGPKOREjtJnX+/mwXeyWbplH31Tk/jNBaO46Ph+dEtO8Lo0kainkBNpoSWb9/LAO+vJ2rSXXp0TuXf6aC6b0J/EOF1rEwkXCjmRZlqes48H38nmk4276ZGSyF0XjmLGxAEkxSvcRMKNQk4kSCty9/PgO9l8lF1A9+QEfn3+SL4/aaB6SYqEMYWcSBNWbTvAg+9k8966XXTtGM9t00Zw1UkD6ZigXx+RcKffUpEGrNl+kIfezebtNfmkdojn1nOHc/XJGXRK1K+NSHuh31aROrLzC3no3WwWfLWTlKQ4fnHOMK45JYPOSRo3UqS9UciJBOw8UMZ/v72eFz/PIzkhjpvPGsoPvzWI1A4KN5H2SiEnUa+wrJLHPvqaJz7ZjM8H131rED+efAxddZ+bSLunkJOoVVntY96Srfzl3Q3sKa5g+ri+3DJlOP27dfS6NBFpJQo5iTrOORau3smf3lrP5t3FnDi4G38/byRj0rt4XZqItDKFnESV5Tn7+MOCtSzL2ccxPTvxxNWZnDmipx51IxKhFHISFbbsLub+hetY8NVOeqQk8oeLj+PS8enExcZ4XZqIhJBCTiLa3uIK/vreBuZm5ZAQF8PPzh7K/zt1MMm6100kKug3XSKSc44nPtnMX97dQHFFFZdPGMDPzx5Kz85JXpcmIm1IIScR6Z01+fzujbWcPqwHvz5/JEN7pXhdkoh4QCEnEaesspp731jD0J6dePzqTOJ13U0kainkJOL838ebyN1byrPXTVLAiUQ57QEkomzbX8ojH27kvON6c/IxaV6XIyIeU8hJRLnvjbUA3HH+KI8rEZFwoJCTiPHZxt288dUOfjz5GPp16eB1OSISBhRyEhEqq3389l+r6d+tAz86bbDX5YhImFDISUSYsyiH7PwifnP+KJLiY70uR0TChEJO2r3dReU8+G42pw3rwTmjenldjoiEEYWctHv3v7WO0opq7rpwlAZaFpHDKOSkXVuRu58XluXxg28NYkiPTl6XIyJhRiEn7ZbP57jrtVX0SEnkJ2ce43U5IhKGFHLSbr24PI8v8w5w+7QRpCTFe12OiIQhhZy0SwdKK/nTW+sYP7ArFx3fz+tyRCRMaexKaZf+8u4G9pZU8I9vT1RnExFpkI7kpN3Jzi/kH4u2MGPiAI7tl+p1OSISxhRy0q445/jt/NV0Sozj1inDvS5HRMKcQk7alTdX7eSzr/dwy5RhdE1O8LocEQlzCjlpN0orqvn9G2sZ2aczV0wa6HU5ItIOKOSk3Xj0w41s21/K3d8eTWyMOpuISNMUctIubN1TwmMfb2L6uL5MHNTN63JEpJ1QyEm7cO8ba4iLMW6fNtLrUkSkHVHISdj7KLuAd9bk85Mzh9I7NcnrckSkHVHISVirqPJx9/zVDEpL5gffyvC6HBFpZxRyEtb+/ulmNu0u5s4LR5EYp4ehikjzKOQkbOUfLOOv723grBE9OWN4T6/LEZF2SCEnYetPb66jstrxmwtGeV2KiLRTCjkJS8tz9vLyF9v4f6cNIiMt2etyRKSdUshJ2Kn2Oe58bTV9UpO48Qw9DFVEWk4hJ2HnuaVbWb39IP913kg6JuhpUCLScgo5CSv7SyqYtXA9Jw7uxgVj+nhdjoi0cwo5CSv//XY2B8uq+O23R+thqCJy1BRyEjbWbD/IM4tzmHniQEb07ux1OSISARRyEhYOPQy1S8cEfn72MK/LEZEIoZCTsDD/y+0s2bKX/zx3OKkd470uR0QihEJOPFdcXsV9C9YyJj2VyzL7e12OiEQQ9c8Wz/3t/Y3kHyznsSvHE6OHoYpIK9KRnHhqU0ERT3yyiUvGp3P8gK5elyMiEUYhJ55xznHP62tIiovlV1NHeF2OiEQghZx45v11u/hwfQE/PXsoPVISvS5HRCKQQk48UVZZzT2vr+GYnp24+uQMr8sRkQiljifiiSc+2UzOnhLm/nAS8bH6v5aIhEZI9y5mNtXM1pvZRjO7rZ7lqWb2LzP70sxWm9m1oaxHwsP2/aU8/P5Gpo7uzbeGpnldjohEsJCFnJnFAo8A04BRwAwzq/v0yxuBNc65scBk4L/NLCFUNUl4uG/BWnzOccf5I70uRUQiXCiP5CYCG51zm5xzFcBzwPQ6bRyQYv6ReDsBe4GqENYkHlv09R5eX7mDGyYPoX+3jl6XIyIRLpQh1w/IrTWdF5hX28PASGA78BXwU+ecr+6KzOxHZrbMzJYVFBSEql4JsapqH3f/azXpXTtw/elDvC5HRKJAKEOuvqErXJ3pc4EVQF9gHPCwmR0x/LxzbrZzLtM5l9mjR4/Wr1TaxNysHNbtLOTX548iKT7W63JEJAqEMuTygNoDEabjP2Kr7VrgZee3EdgM6K7gCLSnqJwH3snm1KFpnDu6l9fliEiUCGXILQWGmtmgQGeS7wHz67TZCpwFYGa9gOHAphDWJB7588L1lFRUc9eFehiqiLSdkN0n55yrMrObgIVALPCkc261mV0fWP4YcC/wlJl9hf/05q+cc7tDVZN4Y2Xefp5flst13xrEMT07eV2OiESRkN4M7pxbACyoM++xWq+3A1NCWYN4yznHXfNXk9YpkZvPGup1OSISZTTUhITUJxt388XW/dw6ZTgpSXoYqoi0LYWchNScRTl0T05g+vF9vS5FRKKQQk5CZvv+Ut5dm8/lE/qTGKdbBkSk7SnkJGTmLdmKA66YNMDrUkQkSinkJCQqqnzMW5LLWSN6kt5Vw3eJiDcUchISC1fvZHdROVeeONDrUkQkiinkJCTmZOUwoFtHThuqYdhExDsKOWl163cWsmTzXq48cQAxMRrdRES8o5CTVjcnawuJcTFcOr5/041FREJIISetqrCsklc+38aFY/vSNVnPvxURbynkpFW9+sU2iiuqmakOJyISBhRy0mqcc8zJymFMeipj+3fxuhwREYWctJ4lm/eSnV+k2wZEJGwo5KTVzMnKIbVDPBeO0TiVIhIeFHLSKnYVlvHWqp1cOj6dDgkap1JEwoNCTlrF80tyqfI5vq9TlSISRhRyctSqqn08u2Qrpw5NY1BastfliIjUUMjJUXtv3S52HCjTbQMiEnYUcnLU5izKoV+XDpw1spfXpYiIHEYhJ0fl64IiPtm4mysmDSBW41SKSJhRyMlReSZrK/GxxmWZGqdSRMKPQk5arKSiin8uz2XasX3okZLodTkiIkdQyEmL/evL7RSWVTHzJHU4EZHwpJCTFnHO8fSiHEb0TiFzYFevyxERqZdCTlpkRe5+Vm8/yJUnDsRMHU5EJDwp5KRF5mTl0Ckxju8c38/rUkREGqSQk2bbW1zB6yt3cPEJ/eiUGOd1OSIiDVLISbO9sCyXiiqfHqkjImFPISfNUu1zPLM4hxMHd2NYrxSvyxERaZRCTprl4+wCcveWMvPEDK9LERFpkkJOmmVOVg49UhKZMlrjVIpI+FPISdBy95bwwfpdzJg4gPhY/dMRkfCnPZUE7ZnFW4kxY8ZEjVMpIu2DQk6CUlZZzQvLcjlnZC/6pHbwuhwRkaAo5CQob67awd7iCo1TKSLtikJOgjJnUQ6DeyRz8pDuXpciIhI0hZw0adW2A3y+dT9XTtI4lSLSvijkpElzs3LoEB/Ld8ene12KiEizKOSkUQdKK3l1xTa+c3xfUjvEe12OiEizKOSkUS8tz6OsUuNUikj7pJCTBjnnmJuVwwkDujC6b6rX5YiINJtCThr02dd72LS7WLcNiEi7pZCTBs1ZlEO35ASmHdvH61JERFpEISf12nGglHfW5nNZZn+S4mO9LkdEpEUUclKveUty8TnH9ycN8LoUEZEWU8jJESqrfcxbspUzhvekf7eOXpcjItJiCjk5wtur8ykoLGembhsQkXZOISdHeHrRFvp368Bpw3p4XYqIyFFRyMlhsvMLWbx5L1dOGkhsjMapFJH2TSEnh5mblUNCXAyXZurBqCLS/inkpEZReRUvf76NC8b0oVtygtfliIgcNYWc1Hj1i20UlVepw4mIRAyFnADfjFN5bL/OjOvfxetyRERahUJOAFiWs491OwuZeaIejCoikUMhJ4B/nMqUpDi+Pbaf16WIiLSaoEPOzJJDWYh4p6CwnDdX7eDS8f3pkKBxKkUkcjQZcmZ2spmtAdYGpsea2f+EvDJpM88v3UplteP7J2qcShGJLMEcyT0InAvsAXDOfQmcFsqipO1UVft4dvFWvnVMGkN6dPK6HBGRVhXU6UrnXG6dWdUhqEU88P66XWw/UKYHo4pIRIoLok2umZ0MODNLAG4mcOpS2r85WTn0SU3irBE9vS5FRKTVBXMkdz1wI9APyAPGAT8OZuVmNtXM1pvZRjO7rYE2k81shZmtNrOPgi1cjt7m3cX8e8Nurpg4gLhYdbQVkcgTzJHccOfc92vPMLNTgE8be5OZxQKPAOfgD8elZjbfObemVpsuwP8AU51zW81MhxNt6JmsHOJijMsnapxKEYlMwfz3/W9BzqtrIrDRObfJOVcBPAdMr9PmCuBl59xWAOfcriDWK62gtKKafy7PY+qxvemZkuR1OSIiIdHgkZyZnQScDPQws1/UWtQZCOZmqn5A7Q4recCkOm2GAfFm9iGQAvzFOfd0PbX8CPgRwIAB6ubeGv61cjsHSis1TqWIRLTGTlcmAJ0CbVJqzT8IXBLEuusbG8rV8/njgbOADsAiM8tyzmUf9ibnZgOzATIzM+uuQ1pgblYOw3p1YuKgbl6XIiISMg2GnHPuI+AjM3vKOZfTgnXnAbUv9qQD2+tps9s5VwwUm9nHwFggGwmZL3P3szLvAPdOH61xKkUkogXT8aTEzP4MjAZqLt44585s4n1LgaFmNgjYBnwP/zW42l4DHjazOPxHjpPw33wuIfT0ohySE2L5zvEap1JEIlswHU+eAdYBg4C7gS34A6xRzrkq4CZgIf776l5wzq02s+vN7PpAm7XAW8BKYAnwuHNuVQu+hwRpX3EF/1q5nYtPSCclKd7rckREQiqYI7nuzrknzOyntU5hBnU/m3NuAbCgzrzH6kz/GfhzsAXL0fnn8lwqqnxcqQ4nIhIFggm5ysCfO8zsfPzX1dJDV5KEis/nmJu1lYmDujG8d0rTbxARaeeCCbnfmVkq8Ev898d1Bn4W0qokJD7eUMDWvSXceu5wr0sREWkTTYacc+71wMsDwBlQM+KJtDNzs3JI65TIuaN7e12KiEibaLDjiZnFmtkMM7vFzI4NzLvAzD4DHm6zCqVV5O4t4b11u5gxsT8JcRqnUkSiQ2NHck/gv89tCfBXM8sBTgJuc8692hbFSeuZt2QrBsyYqBFjRCR6NBZymcAY55zPzJKA3cAxzrmdbVOatJbyqmqeX5rL2SN70bdLB6/LERFpM42dt6pwzvkAnHNlQLYCrn16a9VO9hRX6MGoIhJ1GjuSG2FmKwOvDRgSmDbAOefGhLw6aRVPL8phUFoypwxJ87oUEZE21VjIjWyzKiRkVm8/wPKcffzmglHExGicShGJLo0N0NySQZklzMzN2kpSfAyXnKD790Uk+qgveQQ7WFbJq19sY/rYfqR21DiVIhJ9FHIR7OXleZRWVqvDiYhEraBCzsw6mJnGgmpHnHPMycphXP8uHNsv1etyREQ80WTImdmFwAr8j8TBzMaZ2fxQFyZHZ9GmPXxdUMxMPW1ARKJYMEdyvwUmAvsBnHMrgIzQlSStYW5WDl06xnP+mD5elyIi4plgQq7KOXcg5JVIq8k/WMbC1flcntmfpPhYr8sREfFMMI/aWWVmVwCxZjYUuBn4LLRlydGYt2QrPue4YpLGqRSR6BbMkdxPgNFAOfAs/kfu6HlyYaqy2sezi7dy+rAeDOye7HU5IiKeCuZIbrhz7g7gjlAXI0fvnTX57Cos54+6bUBEJKgjuQfMbJ2Z3Wtmo0NekRyVOYtySO/agdOH9fS6FBERzzUZcs65M4DJQAEw28y+MrNfh7owab6NuwpZtGkP3580kFiNUykiEtzN4M65nc65vwLX479n7s6QViUtMjdrKwmxMVyWqXEqRUQguJvBR5rZb81sFfAw/p6V2ouGmeLyKl5ansf5Y/rQvVOi1+WIiISFYDqe/B2YB0xxzm0PcT3SQq+t2E5heRVXaoQTEZEaTYacc+7EtihEWs45x9OLtjCqT2dOGNDF63JERMJGgyFnZi845y4zs68AV3sRejJ4WPl86z7W7SzkDxcfh5k6nIiIHNLYkdxPA39e0BaFSMs9vSiHlMQ4po/r63UpIiJhpcGOJ865HYGXP3bO5dT+AX7cNuVJU3YXlbPgqx18d3w6HROCucQqIhI9grmF4Jx65k1r7UKkZZ5fmktltdODUUVE6tHYNbkb8B+xDTazlbUWpQCfhrowaVq1z/Hs4q2cckx3hvTo5HU5IiJhp7HzW88CbwJ/AG6rNb/QObc3pFVJUD5Yt4tt+0v5zQUjvS5FRCQsNRZyzjm3xcxurLvAzLop6Lw3JyuHXp0TOXtkL69LEREJS00dyV0ALMd/C0HtvukOGBzCuqQJOXuK+Si7gJ+fPYy42KBGZxMRiToNhpxz7oLAn4ParhwJ1jOLtxIXY3xvYn+vSxERCVvBjF15ipklB15faWYPmJkeOe2hsspqXliWy7mje9Orc5LX5YiIhK1gznM9CpSY2VjgP4EcYE5Iq5JGvb5yB/tLKjVOpYhIE4IJuSrnnAOmA39xzv0F/20E4pE5i7ZwTM9OnDi4m9eliIiEtWBCrtDMbgdmAm+YWSwQH9qypCFf5u7ny7wDzDxxoMapFBFpQjAhdzlQDvzAObcT6Af8OaRVSYPmZuXQMSGWi0/o53UpIiJhr8mQCwTbM0CqmV0AlDnnng55ZXKE/SUVzP9yOxcd34+UJB1Mi4g0JZjelZcBS4BLgcuAxWZ2SagLkyO9uDyP8iqfOpyIiAQpmGHr7wAmOOd2AZhZD+Bd4MVQFiaH8/kcc7NymJDRlZF9OntdjohIuxDMNbmYQwEXsCfI90kr+mTjbrbsKdFRnIhIMwRzJPeWmS0E5gWmLwcWhK4kqc+crBzSOiUw9djeXpciItJuNBlyzrlbzexi4Fv4x6+c7Zx7JeSVSY1t+0t5b20+N0weQmJcrNfliIi0G409T24oMAsYAnwF3OKc29ZWhck35i3eCsCMiRpNTUSkORq7tvYk8DrwXfxPIvhbm1Qkhymvqua5pVs5c0Qv0rt29LocEZF2pbHTlSnOuf8LvF5vZp+3RUFyuLdW7WR3UQUzT1KHExGR5mos5JLM7Hi+eY5ch9rTzjmFXhuYm5VDRveOnHpMmteliIi0O42F3A7ggVrTO2tNO+DMUBUlfmt3HGTpln38+vyRxMRonEoRkeZq7KGpZ7RlIXKkuVk5JMbFcMn4dK9LERFpl3RTd5gqLKvklS+28e2xfenSMcHrckRE2iWFXJh65YttlFRUq8OJiMhRUMiFIecccxblMDY9lTHpXbwuR0Sk3QrmKQRmZlea2Z2B6QFmNjH0pUWvxZv3smFXkcapFBE5SsEcyf0PcBIwIzBdCDwSsoqEOYtySO0Qz4Vj+3pdiohIuxZMyE1yzt0IlAE45/YB6gkRIvkHy1i4eieXZaaTFK9xKkVEjkYwIVdpZrH474079Dw5X0irimLPLcmlyuf4/iSdqhQROVrBhNxfgVeAnmb2e+AT4L6QVhWlKqt9PLskh9OH9SAjLdnrckRE2r1gHrXzjJktB87CP6TXd5xza0NeWRR6b20++QfL+f13dBQnItIaguldOQAoAf4FzAeKA/OaZGZTzWy9mW00s9saaTfBzKrN7JJgC49Ec7Jy6NelA2eM6Ol1KSIiESGYJ4O/gf96nAFJwCBgPTC6sTcFruM9ApwD5AFLzWy+c25NPe3+BCxsdvURZOOuIj7duIdbzx1OrMapFBFpFcGcrjyu9rSZnQD8RxDrnghsdM5tCrzvOWA6sKZOu58ALwETgik4Uj2zOIf4WOPyCf29LkVEJGI0e8STwCN2ggmkfkBurem8wLwaZtYPuAh4rLEVmdmPzGyZmS0rKChoZsXhr6SiiheX53HecX1I65TodTkiIhGjySM5M/tFrckY4AQgmKSp75ybqzP9EPAr51y1WcOn6Jxzs4HZAJmZmXXX0e7NX7GdwrIqZmqEExGRVhXMNbmUWq+r8F+jeymI9+UBtc+9pQPb67TJBJ4LBFwacJ6ZVTnnXg1i/RHBOcfTi3IY0TuF8QO7el2OiEhEaTTkAp1COjnnbm3BupcCQ81sELAN+B5wRe0GzrlBtT7rKeD1aAo4gM+37mfNjoP8/qJjaexoVkREmq/BkDOzOOdcVaCjSbMF3nsT/l6TscCTzrnVZnZ9YHmj1+GixdysHFIS4/jOuH5NNxYRkWZp7EhuCf7rbyvMbD7wT6D40ELn3MtNrdw5twBYUGdeveHmnLsmiHojyp6ict5YuYMrJg0gOTGYM8ciItIcwexZuwF7gDP55n45BzQZctK4F5blUVHt48oTg7q3XkREmqmxkOsZ6Fm5im/C7ZCI6+HY1qp9jmcW53DS4O4c0zOl6TeIiEizNRZysUAngrsVQJrpo+xd5O0r5b/OG+l1KSIiEauxkNvhnLunzSqJMnMW5dAzJZFzRvXyuhQRkYjV2Ign6s8eIlv3lPBhdgEzJg4gPrbZg86IiEiQGtvDntVmVUSZZ5bkEGPGjInqcCIiEkoNhpxzbm9bFhItyiqreWFpLlNG9aJ3apLX5YiIRDSdK2tjb6zcwb6SSo1TKSLSBhRybWxOVg5DeiRz0pDuXpciIhLxFHJt6Ku8A6zI3c/MEwdqnEoRkTagkGtDc7Ny6BAfy8Xj070uRUQkKijk2siBkkpe+3Ib3zm+H52T4r0uR0QkKijk2siLn+dRVqlxKkVE2pJCrg34fI65WTmMH9iV0X1TvS5HRCRqKOTawGdf72Hz7mLdNiAi0sYUcm1gTtYWuiUnMO243l6XIiISVRRyIbZ9fynvrMnn8gn9SYyL9bocEZGoopALsXlLtuKAKzROpYhIm1PIhVBFlY95S3I5c3hP+nfr6HU5IiJRRyEXQgtX72R3UTkzT1KHExERLyjkQmhOVg4DunXktKE9vC5FRCQqKeRCZP3OQpZs3suVJw4gJkbjVIqIeEEhFyJzs3JIiIvh0vH9vS5FRCRqKeRCoKi8ipc/z+PCMX3pmpzgdTkiIlFLIRcCr3yxjeKKanU4ERHxmEKulTnnmLNoC8f1S2VsusapFBHxkkKulS3ZvJfs/CI9GFVEJAwo5FrZnKwcOifFceHYvl6XIiIS9RRyrWhXYRlvrdrJpZn96ZCgcSpFRLymkGtFzy/JpcrnuFKP1BERCQsKudUmMd0AABvmSURBVFZSVe3j2SVbOXVoGoPSkr0uR0REUMi1mvfW7WLHgTI9GFVEJIwo5FrJ3Kwc+qYmceaInl6XIiIiAQq5VrCpoIh/b9jNFZMGEBerTSoiEi60R24FzyzeSnyscdkEjVMpIhJOFHJHqbSimn8uy2XqsX3omZLkdTkiIlKLQu4ozf9yGwfLqtThREQkDCnkjoJzjqcX5TC8VwoTMrp6XY6IiNShkDsKK3L3s3r7Qa48SeNUioiEI4XcUZiTlUOnxDguOr6f16WIiEg9FHIttLe4gtdX7uDiE/rRKTHO63JERKQeCrkW+ueyXCqqfBqnUkQkjCnkWsDnc8xdnMOkQd0Y1ivF63JERKQBCrkW+GhDAbl7S5l5ko7iRETCmUKuBeYuyqFHSiJTRvX2uhQREWmEQq6ZcveW8P76XcyY0J+EOG0+EZFwpr10Mz2zeCsxZsyYNMDrUkREpAkKuWYoq6zmhWW5nD2yJ31SO3hdjoiINEEh1wxvrtrB3uIKrjopw+tSREQkCAq5ZpizKIfBPZI5eUh3r0sREZEgKOSCtGrbAT7fup8rJ2mcShGR9kIhF6RnFueQFB/Dd8ene12KiIgESSEXhAOllbz6xXa+M64fqR3ivS5HRESCpJALwsuf51FaWa1xKkVE2hmFXBOcc8zJyuH4AV04tl+q1+WIiEgzKOSasD6/kE0FxXxvQn+vSxERkWZSyDXhi637AZg0SLcNiIi0Nwq5JnyxdR9dO8YzsHtHr0sREZFmUsg14fOt+zl+QFfdGyci0g4p5BpxoLSSjbuKOGFAF69LERGRFghpyJnZVDNbb2Ybzey2epZ/38xWBn4+M7OxoaynudbvLARgtHpVioi0SyELOTOLBR4BpgGjgBlmNqpOs83A6c65McC9wOxQ1dMS2fn+kBvWK8XjSkREpCVCeSQ3EdjonNvknKsAngOm127gnPvMObcvMJkFhNWYWdn5hXRKjKNvapLXpYiISAuEMuT6Abm1pvMC8xryQ+DN+haY2Y/MbJmZLSsoKGjFEhuXnV/I0F6d1OlERKSdCmXI1ZcMrt6GZmfgD7lf1bfcOTfbOZfpnMvs0aNHK5bYuA35RQzrqVOVIiLtVVwI150H1B4mJB3YXreRmY0BHgemOef2hLCeZtldVM6e4gqG9urkdSkiItJCoTySWwoMNbNBZpYAfA+YX7uBmQ0AXgZmOueyQ1hLs6nTiYhI+xeyIznnXJWZ3QQsBGKBJ51zq83s+sDyx4A7ge7A/wSue1U55zJDVVNzbMgvAmB4b4WciEh7FcrTlTjnFgAL6sx7rNbr64DrQllDS2XnF9I5KY6eKYlelyIiIi2kEU8asCG/iGG9UtSzUkSkHVPI1cM5x/r8QobqepyISLumkKtHQWE5B0orGa6elSIi7ZpCrh7ZgU4n6lkpItK+KeTqcej2AZ2uFBFp3xRy9cjOL6Rrx3jSOiV4XYqIiBwFhVw9svML1bNSRCQCKOTqcM7V3D4gIiLtm0Kujp0Hyygsr2KYelaKiLR7Crk6DvWsVKcTEZH2TyFXR/ZODcwsIhIpFHJ1ZOcXktYpkW7J6lkpItLeKeTqyN5VpOtxIiIRQiFXi8/n2Bi4fUBERNo/hVwt2/aXUlxRrZATEYkQCrlaNuw61OlEpytFRCKBQq4W3T4gIhJZFHK1ZOcX0qtzIqkd4r0uRUREWoFCrpZsdToREYkoCrkAn8+xcZfGrBQRiSQKuYDcfSWUVfrU6UREJIIo5ALU6UREJPIo5AJqngbeU0dyIiKRQiEXkJ1fSL8uHUhJUs9KEZFIoZALyM4vYqiux4mIRBSFHFBV7ePrAvWsFBGJNAo5IGdvCRVVPl2PExGJMAo5YEOg08nw3jqSExGJJAo5vrl94BgdyYmIRBSFHP6elf27daBjQpzXpYiISCtSyBEYs7KnTlWKiESaqA+5ymofm3cXM0zX40REIk7Uh9yW3cVUVjuNWSkiEoGiPuRqxqzU6UoRkYijkMsvJMbUs1JEJBIp5PILGdg9maT4WK9LERGRVqaQyy/USCciIhEqqkOuvKqaLXtKNGaliEiEiuqQ27y7mGqf09MHREQiVFSH3PqdGrNSRCSSRXXIbcgvIjbGGJSW7HUpIiISAlEdctn5hWR070hinHpWiohEoqgOuQ279KBUEZFIFrUhV1ZZzZY9xQxVyImIRKyoDbmNu4pwDoYr5EREIlbUhtyGXf6elRqYWUQkckVtyGXnFxEfa2SoZ6WISMSK3pDbWcigtGTiY6N2E4iIRLyo3cNn7ypUz0oRkQgXlSFXUlFF7t5ShZyISISLypDbuMv/oFR1OhERiWxRGXKHxqzUPXIiIpEtKkPu64Ji4mONgd06el2KiIiEUFSGXHF5FSlJ8cSpZ6WISESLyr18RZWPBAWciEjEi8o9fWW1j/g487oMEREJsagMufJqHcmJiESDqNzTV1T5NNKJiEgUiPO6AC9UVvtIjFPISWhVVlaSl5dHWVmZ16WIRIykpCTS09OJj48Pqn1UhpyO5KQt5OXlkZKSQkZGBma6BixytJxz7Nmzh7y8PAYNGhTUe0K6pzezqWa23sw2mtlt9Sw3M/trYPlKMzshlPUcUlntI0FHchJiZWVldO/eXQEn0krMjO7duzfr7EjI9vRmFgs8AkwDRgEzzGxUnWbTgKGBnx8Bj4aqntp0JCdtRQEn0rqa+zsVyj39RGCjc26Tc64CeA6YXqfNdOBp55cFdDGzPiGsCQCfg9gY7XxERCJdKEOuH5BbazovMK+5bUSkhWJjYxk3bhyjR49m7NixPPDAA/h8vhat68477+Tdd99tcPljjz3G008/3ez1Lly4kHHjxjFu3Dg6derE8OHDGTduHFdddVWL6qxt1qxZjBgxgmOPPZaxY8fW1Dd58mSWLVt21OsHWLZsGTfffDMA5eXlnH322YwbN47nn3+e6667jjVr1hzV+h966KHDtmtVVRVpaWncfvvth7XLyMhg9+7dNdMffvghF1xwQc30m2++SWZmJiNHjmTEiBHccsstR1UXwPLlyznuuOM45phjuPnmm3HOHdGmsrKSq6++muOOO46RI0fyhz/84Yg23/72tzn22GNrph9++GH+/ve/H3V9gP9CXih+gEuBx2tNzwT+VqfNG8C3ak2/B4yvZ10/ApYBywYMGOCO1pe5+9ya7QeOej0ijVmzZo3XJbjk5OSa1/n5+e6ss85yd955p4cVNe700093S5cuPWJ+VVVVs9f16KOPuilTprgDB/y/6/v373dPPfVUo59ztBYtWuROO+20Fr+/7vesrKx0xx13nKusrKyZ98Ybb7iTTz7ZDR482Pl8vpr5AwcOdAUFBTXTH3zwgTv//POdc8599dVXbvDgwW7t2rU1633kkUdaXOchEyZMcJ999pnz+Xxu6tSpbsGCBUe0eeaZZ9zll1/unHOuuLjYDRw40G3evLlm+UsvveRmzJjhRo8eXTOvuLjYjRs3rsHPre93C1jm6smiUPauzAP615pOB7a3oA3OudnAbIDMzMwj/6vQTGPSuxztKkSa5e5/rWbN9oOtus5RfTtz14Wjg27fs2dPZs+ezYQJE/jtb3+Lz+fjtttu48MPP6S8vJwbb7yR//iP/wDg/vvvZ86cOcTExDBt2jT++Mc/cs0113DBBRdwySWXcNtttzF//nzi4uKYMmUKs2bN4re//S2dOnXilltuYcWKFVx//fWUlJQwZMgQnnzySbp27crkyZOZNGkSH3zwAfv37+eJJ57g1FNPrbfejIwMfvCDH/D2229z00030a1bN+666y7Ky8sZMmQIf//73+nUqRPLly/nF7/4BUVFRaSlpfHUU0/Rp08f7rvvPj744AM6d+4MQGpqKldfffURn3PDDTewdOlSSktLueSSS7j77rsB6v2O//znP7n77ruJjY0lNTWVjz/+mA8//JBZs2bx5JNPcuWVV1JQUMC4ceN46aWX+OEPf8isWbPIzMzk7bffrrf+ut/ze9/7Xk1t77//PieccAJxcd/squfNm8dPf/pTHn30UbKysjjppJOa/Lu///77ueOOOxgxYgQAcXFx/PjHPw7yX079duzYwcGDB2s+/6qrruLVV19l2rRph7UzM4qLi6mqqqK0tJSEhISav5OioiIeeOABZs+ezWWXXVbzno4dO5KRkcGSJUuYOHHiUdUZypBbCgw1s0HANuB7wBV12swHbjKz54BJwAHn3I4Q1iQS1QYPHozP52PXrl289tprpKamsnTpUsrLyznllFOYMmUK69at49VXX2Xx4sV07NiRvXv3HraOvXv38sorr7Bu3TrMjP379x/xOVdddRV/+9vfOP3007nzzju5++67eeihhwD/6bYlS5awYMEC7r777kZPgSYlJfHJJ5+we/duLr74Yt59912Sk5P505/+xAMPPMDtt9/OT37yE1577TV69OjB888/zx133MFf/vIXCgsLGTJkSJPb5Pe//z3dunWjurqas846i5UrV5Kenl7vd7znnntYuHAh/fr1O+J79+zZk8cff5xZs2bx+uuvH7Zs9+7d/O53vzui/jvvvPOw71nXp59+yvjx42umS0tLee+99/jf//1f9u/fz7x584IKuVWrVvHLX/6yyXYffPABP//5z4+Y37FjRz777LPD5m3bto309PSa6fT0dLZt23bEey+55BJee+01+vTpQ0lJCQ8++CDdunUD4De/+Q2//OUv6djxyCfCZGZm8u9//zt8Q845V2VmNwELgVjgSefcajO7PrD8MWABcB6wESgBrg1VPSJeas4RV6i5wHWTt99+m5UrV/Liiy8CcODAATZs2MC7777LtddeW7PjObRDOqRz584kJSVx3XXXcf755x923efQevbv38/pp58OwNVXX82ll15as/ziiy8GYPz48WzZsqXRWi+//HIAsrKyWLNmDaeccgoAFRUVnHTSSaxfv55Vq1ZxzjnnAFBdXU2fPn1wzgXdC++FF15g9uzZVFVVsWPHDtasWcOoUaPq/Y6nnHIK11xzDZdddlnN9whGQ/XX/Z517dixg5EjR9ZMv/7665xxxhl07NiR7373u9x77708+OCDxMbG1vt9m9sT8YwzzmDFihVBtT3076ipz1uyZAmxsbFs376dffv2ceqpp3L22Wdz8OBBNm7cyIMPPljvv4OePXuybt26ZtVfn5DeDO6cW4A/yGrPe6zWawfcGMoaROQbmzZtIjY2lp49e+Kc429/+xvnnnvuYW3eeuutRneOcXFxLFmyhPfee4/nnnuOhx9+mPfffz/oGhITEwF/p5iqqqpG2yYnJwP+Heo555zDvHnzDlv+1VdfMXr0aBYtWlTvezdt2sTgwYMbXP/mzZuZNWsWS5cupWvXrlxzzTWUlZU1+B0fe+wxFi9ezBtvvMG4ceOaFQj11V/3e9bVoUOHw+4JmzdvHp9++ikZGRkA7Nmzhw8++ICzzz6b7t27s2/fPtLS0gD/Efeh16NHj2b58uWMHTu20TqbcySXnp5OXl5ezXReXh59+/Y94r3PPvssU6dOJT4+np49e3LKKaewbNky9uzZw/Lly8nIyKCqqopdu3YxefJkPvzwQ8B/n2mHDh0arTcYullMJEoUFBRw/fXXc9NNN2FmnHvuuTz66KNUVlYCkJ2dTXFxMVOmTOHJJ5+kpKQE4IjTlUVFRRw4cIDzzjuPhx566IgdfWpqKl27duXf//43AHPmzKk5qmupE088kU8//ZSNGzcCUFJSQnZ2NsOHD6egoKAm5CorK1m9ejUAt99+OzfeeCMHD/qvhR48eJDZs2cftt6DBw+SnJxMamoq+fn5vPnmm41+x6+//ppJkyZxzz33kJaWRm5uLsFoqP6mjBw5suY9Bw8e5JNPPmHr1q1s2bKFLVu28Mgjj9QE5+TJk5kzZw7gP6KdO3cuZ5xxBgC33nor9913X81n+nw+HnjggSM+79CRXN2fugEH0KdPH1JSUsjKysI5x9NPP8306XXvEoMBAwbw/vvv45yjuLiYrKwsRowYwQ033MD27dvZsmULn3zyCcOGDasJOPD/e6zd47KlonJYL5FoUVpayrhx46isrCQuLo6ZM2fyi1/8AoDrrruOLVu2cMIJJ+Cco0ePHrz66qtMnTqVFStWkJmZSUJCAueddx733XdfzToLCwuZPn06ZWVlOOd48MEHj/jcf/zjHzUdTwYPHnzU3cF79OjBU089xYwZMygvLwfgd7/7HcOGDePFF1/k5ptv5sCBA1RVVfGzn/2M0aNHc8MNN1BUVMSECROIj48nPj7+iOtSY8eO5fjjj2f06NEMHjy45nRiQ9/x1ltvZcOGDTjnOOussxg7diwfffTRUdXfmGnTpjFz5kwAXn75Zc4888yaI2GA6dOn85//+Z+Ul5fzm9/8hhtuuIGxY8finGPq1KlceeWVAIwZM4aHHnqIGTNmUFJSgplx/vnnB7PpG/Xoo49yzTXXUFpayrRp02o6ncyfP59ly5Zxzz33cOONN3Lttddy7LHH4pzj2muvZcyYMU2u+9NPP+Wuu+466hqtvvOq4SwzM9O11v0tIqG0du3aw66niLTERRddxP3338/QoUO9LqXNfPHFFzzwwAM1R6Z11fe7ZWbLnXOZddvqdKWISBj74x//yI4d0dXpfPfu3dx7772tsi6drhQRCWPDhw9n+PDhXpfRpg71lm0NOpITCaH2djlAJNw193dKIScSIklJSezZs0dBJ9JKXOB5cklJSUG/R6crRULk0H1EBQUFXpciEjEOPRk8WAo5kRCJj48P+unFIhIaOl0pIiIRSyEnIiIRSyEnIiIRq92NeGJmBUBOK6wqDdjdZKvopG3TMG2bhmnbNEzbpmGttW0GOud61J3Z7kKutZjZsvqGgBFtm8Zo2zRM26Zh2jYNC/W20elKERGJWAo5ERGJWNEccrObbhK1tG0apm3TMG2bhmnbNCyk2yZqr8mJiEjki+YjORERiXAKORERiVgRH3JmNtXM1pvZRjO7rZ7lZmZ/DSxfaWYneFGnF4LYNt8PbJOVZvaZmY31ok4vNLVtarWbYGbVZnZJW9bnpWC2jZlNNrMVZrbazD5q6xq9EsTvVKqZ/cvMvgxsm2u9qLOtmdmTZrbLzFY1sDx0+2HnXMT+ALHA18BgIAH4EhhVp815wJuAAScCi72uO4y2zclA18Drado29bZ7H1gAXOJ13eGybYAuwBpgQGC6p9d1h9G2+S/gT4HXPYC9QILXtbfBtjkNOAFY1cDykO2HI/1IbiKw0Tm3yTlXATwHTK/TZjrwtPPLArqYWZ+2LtQDTW4b59xnzrl9gcksIPjnW7Rvwfy7AfgJ8BKwqy2L81gw2+YK4GXn3FYA51y0bJ9gto0DUszMgE74Q66qbctse865j/F/14aEbD8c6SHXD8itNZ0XmNfcNpGoud/7h/j/pxUNmtw2ZtYPuAh4rA3rCgfB/LsZBnQ1sw/NbLmZXdVm1XkrmG3zMDAS2A58BfzUOedrm/LCWsj2w5H+PDmrZ17deyaCaROJgv7eZnYG/pD7VkgrCh/BbJuHgF8556r9/ymPGsFsmzhgPHAW0AFYZGZZzrnsUBfnsWC2zbnACuBMYAjwjpn92zl3MNTFhbmQ7YcjPeTygP61ptPx/w+quW0iUVDf28zGAI8D05xze9qoNq8Fs20ygecCAZcGnGdmVc65V9umRM8E+zu12zlXDBSb2cfAWCDSQy6YbXMt8EfnvxC10cw2AyOAJW1TYtgK2X440k9XLgWGmtkgM0sAvgfMr9NmPnBVoHfPicAB59yOti7UA01uGzMbALwMzIyC/4XX1uS2cc4Ncs5lOOcygBeBH0dBwEFwv1OvAaeaWZyZdQQmAWvbuE4vBLNttuI/wsXMegHDgU1tWmV4Ctl+OKKP5JxzVWZ2E7AQf8+nJ51zq83s+sDyx/D3jDsP2AiU4P+fVsQLctvcCXQH/idwxFLlomAk9SC3TVQKZts459aa2VvASsAHPO6cq7freCQJ8t/NvcBTZvYV/lN0v3LORfwjeMxsHjAZSDOzPOAuIB5Cvx/WsF4iIhKxIv10pYiIRDGFnIiIRCyFnIiIRCyFnIiIRCyFnIiIRCyFnAgQeJLAilo/GY20LWqFz3vKzDYHPutzMzupBet43MxGBV7/V51lnx1tjYH1HNouqwKj53dpov04MzuvNT5bpDXoFgIR/MHlnOvU2m0bWcdTwOvOuRfNbAowyzk35ijWd9Q1NbVeM/sHkO2c+30j7a8BMp1zN7V2LSItoSM5kXqYWSczey9wlPWVmR3xFAIz62NmH9c60jk1MH+KmS0KvPefZtZU+HwMHBN47y8C61plZj8LzEs2szcCzyBbZWaXB+Z/aGaZZvZHoEOgjmcCy4oCfz5f+8gqcAT5XTOLNbM/m9lS8z+/6z+C2CyLCAyaa2YTzf+MwS8Cfw4PjPJxD3B5oJbLA7U/GficL+rbjiIh5fVzhvSjn3D4AarxD5y7AngF/2hAnQPL0vCPxHDozEdR4M9fAncEXscCKYG2HwPJgfm/Au6s5/OeIvAMOuBSYDH+QY2/ApLxP4ZlNXA88F3g/2q9NzXw54f4j5pqaqrV5lCNFwH/CLxOwD/SewfgR8CvA/MTgWXAoHrqLKr1/f4JTA1MdwbiAq/PBl4KvL4GeLjW++8Drgy87oJ//Mpkr/++9RM9PxE9rJdIM5Q658YdmjCzeOA+MzsN/9BU/YBewM5a71kKPBlo+6pzboWZnQ6MAj4NDIWWgP8IqD5/NrNfAwX4n/JwFvCK8w9sjJm9DJwKvAXMMrM/4T/F+e9mfK83gb+aWSIwFfjYOVcaOEU6xr55onkqMBTYXOf9HcxsBZABLAfeqdX+H2Y2FP9o8fENfP4U4NtmdktgOgkYQHSMZSlhQCEnUr/v439y83jnXKWZbcG/g67hnPs4EILnA3PM7M/APuAd59yMID7jVufci4cmzOzs+ho557LNbDz+sf3+YGZvO+fuCeZLOOfKzOxD/I94uRyYd+jjgJ845xY2sYpS59w4M0sFXgduBP6KfwzGD5xzFwU66XzYwPsN+K5zbn0w9Yq0Nl2TE6lfKrArEHBnAAPrNjCzgYE2/wc8AZyA/wnqp5jZoWtsHc1sWJCf+THwncB7kvGfavy3mfUFSpxzc4FZgc+pqzJwRFmf5/APeHsq/sGDCfx5w6H3mNmwwGfWyzl3ALgZuCXwnlRgW2DxNbWaFuI/bXvIQuAnFjisNbPjG/oMkVBQyInU7xkg08yW4T+qW1dPm8nACjP7Av91s7845wrw7/TnmdlK/KE3IpgPdM59jv9a3RL81+ged859ARwHLAmcNrwD+F09b58NrDzU8aSOt4HTgHedcxWBeY8Da4DPzWwV8L80cWYnUMuX+B8hcz/+o8pP8V+vO+QDYNShjif4j/jiA7WtCkyLtBndQiAiIhFLR3IiIhKxFHIiIhKxFHIiIhKxFHIiIhKxFHIiIhKxFHIiIhKxFHIiIhKx/j9EpDXtn9UHYwAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<Figure size 504x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# roc plot\n",
    "from sklearn.metrics import plot_roc_curve\n",
    "fig,ax = plt.subplots(figsize=(7,7))\n",
    "plot_roc_curve(clf, X_test, y_test, ax = ax)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}