metro_project_irina_baeva-v1.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np \n",
    "import pandas as pd\n",
    "from urllib import request\n",
    "\n",
    "%matplotlib inline\n",
    "pd.options.mode.chained_assignment = None #to avoid SettingWithCopyWarning\n",
    "import matplotlib.pyplot as plt\n",
    "\n",
    "import seaborn as sns \n",
    "\n",
    "from dateutil.parser import parse\n",
    "from datetime import date\n",
    "import calendar\n",
    "import matplotlib.dates as mdates\n",
    "import matplotlib.patches as mpatches\n",
    "\n",
    "import scipy.stats as st\n",
    "from scipy.stats import norm"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Exploratory analysis of sales and shrinkage"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Abstract\n",
    "Current report contains work with the dataset provided by \"CODE\" university partner \"Metronom\". \"Metronom\" is a part of the huge German retailer company \"Metro\". The data contains information from the one German store. Time interval of the data set - is a period between 1st to 31th of January 2020. Data could not be shared openly due to non-disclosure agreement. German is a main language of the data set.  \n",
    "\n",
    "The main question of the report is how sales and shrinkage behave during the current period depending on the weekday, purchasing areas of products, freshness. Results were achieved by exploratory data analysis, visualizations with Python libraries and statistical methods. By exploring the target variables, we found that sales drop on Fridays, on contrary shrinkage takes the biggest part on Fridays. All products could be divided by the two main categories depending on their freshness: normal fresh (fresh) and ultrafresh. 58,9% of unique products belong to normal fresh and 41,1% belong to the ultrafresh (almost equal parts), but shrinkage for January 2020 contains a bigger part of the normal fresh products, then the ultrafresh products. Also, we explored products with the highest shrinkage and sales. It was proved that shrinkage and sales does not have any correlation. Sales variable has lognormal distribution and 75% of the sales are between 1 and 54 euros. Having these results would be important for further study of other time intervals and dependencies."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Introduction\n",
    "\n",
    "How do we find the main patterns of sales and shrinkage? Is there a relationship between shrinkage and sales? What are the products with the highest shrinkage?  We hypothesised that we could explore relationships between categorical and numerical variables and find dependencies. Following the step by step analyse process, we explored trends of sales and shrinkage by a day and by  a weekday and found that while sales drops on Fridays, shrinkage stays around 42% out of total.  Fresh and ultrafresh products take 56% and 44% of total sales and 73,6% and 26,4% of total shrinkage respectively. The top 3 products with the highest shrinkage take around 7% out of the total shrinkage and make only 0,58% of total sales which could be considered as the reason to give up on selling these products. Exploration of correlation metrics showed that there is no correlation between our target variables. In addition, we found that the most suitable theoretical distribution of the sales is positively skewed lognormal distribution and 75% of the sales values lay in the interval between 1 and 54 euros.\n",
    "\n",
    "In order to achieve the goal we worked with different categories of data and used exploratory data analysis and statistical methods. Since the data was row - cleaning, consulting with the company data experts and finding a way to work with the missing values took a big part of the results. The structure of the analysis process is as follows: loading and preprocessing the data, cleaning the data, exploring the data. Since shrinkage is a very complex variable which depends on human factors, there is not enough information in the current data to tackle the problem of decreasing shrinkage so it could be considered as a target for further analysis.\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Data, materials and Methods\n",
    "\n",
    "**Data**\n",
    "\n",
    "The data set was provided by one of the biggest german retail companies __[\"METRO\"](https://www.metro.de/)__ . It contains information about products for the one german department store. Data could not be shared openly due to non-disclosure agreement\n",
    "\n",
    "*Time interval of the data set* -  period between 1st to 31th of January 2020. \n",
    "\n",
    "*German* is a main language of the data set. Names of the columns were translated for more convenient work in the Loading and preprocessing the data. \n",
    "\n",
    "Translation of categories is as following: *ACM FLEISCH* - meat category, *ACM BACKWAREN* - bakery products category, *ACM FRISCHFISCH* - fresh fish category, *ACM MOLKEREIPRODUKTE* - dairy products category, *ACM KAESE* - cheese products category, *ACM FEINKOST* - delicatessen category, *ACM GOURVENIENCE* - the own brand of \"Metro\" category, *ACM WURST* - sausage category, *ACM OBST* - fruits category, *ACM GEMUESE* - vegetables category.\n",
    "\n",
    "Each purchasing area is a part of 2 big categories divided by freshness. We have an information from \"Metro\" with following description:  ACM FLEISCH, ACM FRISCHFISCH, ACM GEMUESE, ACM BLUMEN, ACM OBST belong to \"ultrafresh\" category (5 purchasing areas) and rest of it belong to normal fresh. \n",
    "\n",
    "**Keywords**:\n",
    "\n",
    "*Depo* is a B2B shop for the small businesses (HORECA): shops, cafes, restaurants.\n",
    "\n",
    "*Shrinkage* is a waste of products due to different reasons such as theft, damage, spoilage.\n",
    "\n",
    "*Freshness* is an attribute of the perceived quality of food.\n",
    "\n",
    "\n",
    "**Reqired libraries:**\n",
    "- __[Numpy](https://www.numpy.org/)__ ,\n",
    "- __[Pandas](https://pandas.pydata.org/)__ ,\n",
    "- __[Seaborn](https://seaborn.pydata.org/)__ ,\n",
    "- __[Matplotlib](https://matplotlib.org/)__  ,\n",
    "- __[SciPy](https://www.scipy.org/)__ ,\n",
    "- __[SciPy.stats](https://docs.scipy.org/doc/scipy/reference/stats.html)__ .\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Analysis"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### i. Loading and preprocessing the data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "filepath = \"./data_metro.csv\"\n",
    "df = pd.read_csv(filepath, encoding = \"ISO-8859-1\", error_bad_lines=False,low_memory=False, sep=\"\\t\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Current dataset contains 13 columns, most of them are stored as a string format. A few target columns have missing values.\n",
    "We need to start cleaning the data."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['store_id', 'store_desc', 'CATMAN_BUY_DOMAIN_ID',\n",
       "       'CATMAN_BUY_DOMAIN_DESC', 'PCG_MAIN_CAT_ID', 'PCG_CAT_ID',\n",
       "       'PCG_CAT_DESC', 'PCG_SUB_CAT_ID', 'PCG_SUB_CAT_DESC', 'ART_NO',\n",
       "       'ART_NAME', 'Bv NNEK', 'Bv Stk', 'Bestand Ist NNEK',\n",
       "       'Bestand Ist Stück', 'WA Stück', 'Umsatz VK Netto', 'date_of_day',\n",
       "       'DAY_DESC'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.columns"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Since we have raw data, column names need clarification and should be renamed to more accurately described."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>store_id</th>\n",
       "      <th>store_desc</th>\n",
       "      <th>purchasing_area_id</th>\n",
       "      <th>purchasing_area_description</th>\n",
       "      <th>main_product_group_id</th>\n",
       "      <th>product_group_category_id</th>\n",
       "      <th>product_group_category_description</th>\n",
       "      <th>product_subgroup_category_id</th>\n",
       "      <th>product_subgroup_category_description</th>\n",
       "      <th>article_number</th>\n",
       "      <th>article_name</th>\n",
       "      <th>daily_shrinkage_in_euros</th>\n",
       "      <th>daily_shrinkage_in_pieces</th>\n",
       "      <th>daily_stock_in_euros</th>\n",
       "      <th>daily_stock_in_pieces</th>\n",
       "      <th>daily_sales_in_pieces</th>\n",
       "      <th>daily_sales_in_euros</th>\n",
       "      <th>date_of_day</th>\n",
       "      <th>DAY_DESC</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>223946</td>\n",
       "      <td>579</td>\n",
       "      <td>LIEFERDEPOT BERLIN</td>\n",
       "      <td>53</td>\n",
       "      <td>ACM FEINKOST</td>\n",
       "      <td>945</td>\n",
       "      <td>46</td>\n",
       "      <td>Beilagen</td>\n",
       "      <td>20</td>\n",
       "      <td>Schupfnudeln</td>\n",
       "      <td>133372</td>\n",
       "      <td>500g SCHUPFNUDELN LANDMANNS</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>31.01.20</td>\n",
       "      <td>31.01.20 Fre</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        store_id          store_desc  purchasing_area_id  \\\n",
       "223946       579  LIEFERDEPOT BERLIN                  53   \n",
       "\n",
       "       purchasing_area_description  main_product_group_id  \\\n",
       "223946                ACM FEINKOST                    945   \n",
       "\n",
       "        product_group_category_id product_group_category_description  \\\n",
       "223946                         46                           Beilagen   \n",
       "\n",
       "        product_subgroup_category_id product_subgroup_category_description  \\\n",
       "223946                            20                          Schupfnudeln   \n",
       "\n",
       "        article_number                 article_name daily_shrinkage_in_euros  \\\n",
       "223946          133372  500g SCHUPFNUDELN LANDMANNS                      NaN   \n",
       "\n",
       "        daily_shrinkage_in_pieces  daily_stock_in_euros  \\\n",
       "223946                        NaN                   0.0   \n",
       "\n",
       "        daily_stock_in_pieces  daily_sales_in_pieces  daily_sales_in_euros  \\\n",
       "223946                    0.0                    NaN                   NaN   \n",
       "\n",
       "       date_of_day      DAY_DESC  \n",
       "223946    31.01.20  31.01.20 Fre  "
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df.rename(columns={\"CATMAN_BUY_DOMAIN_ID\": \"purchasing_area_id\", \n",
    "                   \"CATMAN_BUY_DOMAIN_DESC\" : \"purchasing_area_description\", \n",
    "                   \"PCG_MAIN_CAT_ID\" : \"main_product_group_id\",\n",
    "                   \"PCG_CAT_ID\" : \"product_group_category_id\",\n",
    "                   \"PCG_CAT_DESC\" : \"product_group_category_description\",\n",
    "                   \"PCG_SUB_CAT_ID\" : \"product_subgroup_category_id\",\n",
    "                   \"PCG_SUB_CAT_DESC\" : \"product_subgroup_category_description\",\n",
    "                            \"ART_NO\": \"article_number\",\n",
    "                            \"ART_NAME\":\"article_name\",\n",
    "                   \"Bv NNEK\" : \"daily_shrinkage_in_euros\",\n",
    "                   \"Bv Stk\" : \"daily_shrinkage_in_pieces\",\n",
    "                   \"Bestand Ist NNEK\": \"daily_stock_in_euros\",\n",
    "                   \"Bestand Ist Stück\": \"daily_stock_in_pieces\",\n",
    "                   \"WA Stück\": \"daily_sales_in_pieces\",\n",
    "                   \"Umsatz VK Netto\" : \"daily_sales_in_euros\"\n",
    "                   \n",
    "                  })\n",
    "df.sample(1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The data dictionary provided with the dataset is as the following:\n",
    "\n",
    "**store_id and store_desc** - number and name of the department store, we have got a data from one store; \n",
    "\n",
    "**purchasing_area_id** and **purchasing_area_description** - id and name of the main categories of products (11);\n",
    "\n",
    "**product_group_category_id** and **product_group_category_description** - id and name of detailed categories of products (179);\n",
    "\n",
    "**product_subgroup_category_id** and **product_subgroup_category_description** - id and name of  subcategories of products (773);\n",
    "\n",
    "**article_number** and **article_name**- number (23845) and name (23003) of each product;\n",
    "\n",
    "**daily_shrinkage_in_euros** and **daily_shrinkage_in_pieces** - shrinkage of each product in euros and in pieces;\n",
    "\n",
    "**daily_stock_in_euros** and **daily_stock_in_pieces** - daily stock of each product in euros and in pieces;\n",
    "\n",
    "**daily_sales_in_euros** and **daily_sales_in_pieces** - daily sales of each product in euros and in pieces;\n",
    "\n",
    "**date_of_day** - date (YYYY-MM-DD)\n",
    "\n",
    "**DAY_DESC** - date in a different format (DD.MM.YY Weekday)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We are going to study different product categories to familiarize ourselves with them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "11"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df[\"purchasing_area_description\"].unique().tolist())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are 11 purchasing areas of products."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "179"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df[\"product_group_category_description\"].unique())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are 179 product categories."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "773"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df[\"product_subgroup_category_description\"].unique().tolist())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are 773 product subcategories. \n",
    "\n",
    "Is there the same amount of article numbers as article names?\n",
    "First, there are 738671 rows in total. Now let's check how many unique article numbers and related products in the data set."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "23845"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "array_of_art_num = df[\"article_number\"].unique().tolist()\n",
    "len(array_of_art_num)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we check number of names of the product."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "23003"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "array_of_art_name = df[\"article_name\"].unique().tolist()\n",
    "len(array_of_art_name)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We got less product names then article numbers. This question was clarified with the company data analyst. It is happening because sometimes the same products has the different article numbers (different taste, flavour and so on). For current report we will use article name."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## ii. Cleaning the data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "First, we explore types of data to find obvious issues."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "store_id                                   int64\n",
       "store_desc                                object\n",
       "purchasing_area_id                         int64\n",
       "purchasing_area_description               object\n",
       "main_product_group_id                      int64\n",
       "product_group_category_id                  int64\n",
       "product_group_category_description        object\n",
       "product_subgroup_category_id               int64\n",
       "product_subgroup_category_description     object\n",
       "article_number                             int64\n",
       "article_name                              object\n",
       "daily_shrinkage_in_euros                  object\n",
       "daily_shrinkage_in_pieces                float64\n",
       "daily_stock_in_euros                     float64\n",
       "daily_stock_in_pieces                    float64\n",
       "daily_sales_in_pieces                    float64\n",
       "daily_sales_in_euros                     float64\n",
       "date_of_day                               object\n",
       "DAY_DESC                                  object\n",
       "dtype: object"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Exploring the types of data frame, we identified several problems that need to be solved in order to continue working with data:\n",
    "* daily_shrinkage_in_euros has an object type;\n",
    "* categorical columns have an object type;\n",
    "* date_of_day has an object type."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "dtype('float64')"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"daily_shrinkage_in_euros\"].replace(',','.',inplace = True, regex = True)\n",
    "df[[\"daily_shrinkage_in_euros\"]] = df[[\"daily_shrinkage_in_euros\"]].apply(pd.to_numeric)\n",
    "df[\"daily_shrinkage_in_euros\"].dtype"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "for col in [\"purchasing_area_description\", \"product_group_category_description\", \"product_subgroup_category_description\", \"article_name\"]:\n",
    "    df[col] = df[col].astype(\"category\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The date is represented by two columns “date_of_day” and “DAY_DESC”. We decided to use the date_of_day column and, in addition to changing the type, extract additional information about the days of the week and save it in a new column."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"date_of_day\"]=pd.to_datetime(df[\"date_of_day\"], format=\"%d.%m.%y\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, when the column \"date_of_day\" has a datetime type, we can extract information about weekdays. This is going to be an important column for the explorational part of the report later."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "store_id                                          int64\n",
       "store_desc                                       object\n",
       "purchasing_area_id                                int64\n",
       "purchasing_area_description                    category\n",
       "main_product_group_id                             int64\n",
       "product_group_category_id                         int64\n",
       "product_group_category_description             category\n",
       "product_subgroup_category_id                      int64\n",
       "product_subgroup_category_description          category\n",
       "article_number                                    int64\n",
       "article_name                                   category\n",
       "daily_shrinkage_in_euros                        float64\n",
       "daily_shrinkage_in_pieces                       float64\n",
       "daily_stock_in_euros                            float64\n",
       "daily_stock_in_pieces                           float64\n",
       "daily_sales_in_pieces                           float64\n",
       "daily_sales_in_euros                            float64\n",
       "date_of_day                              datetime64[ns]\n",
       "DAY_DESC                                         object\n",
       "weekday                                        category\n",
       "dtype: object"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df['weekday'] = df[\"date_of_day\"].dt.dayofweek.apply( lambda x: calendar.day_name[x])\n",
    "df['weekday'] = df['weekday'].astype('category')\n",
    "df.dtypes"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As a result of this part, we fixed types of data, but we have some data which is not going to be useful for our report."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Exploring the null values and redundant data."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "There are columns with a redundant data which is not useful for our project. We can easily drop following colomns: \n",
    "- \"store_id\" and \"store_desc\", since we have data for only one store;\n",
    "- \"DAY_DESC\", since we have enough information about date;\n",
    "- purchasing_area_id, \"main_product_group_id\", \"product_group_category_id\", \"product_subgroup_category_id\", \"article_number\" columns do not have usefull information for the current report."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = df.drop([\"store_id\",\"store_desc\",\"DAY_DESC\", \"article_number\", \"purchasing_area_id\",\"main_product_group_id\", \"product_group_category_id\",\"product_subgroup_category_id\"], axis = 1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>purchasing_area_description</th>\n",
       "      <th>product_group_category_description</th>\n",
       "      <th>product_subgroup_category_description</th>\n",
       "      <th>article_name</th>\n",
       "      <th>daily_shrinkage_in_euros</th>\n",
       "      <th>daily_shrinkage_in_pieces</th>\n",
       "      <th>daily_stock_in_euros</th>\n",
       "      <th>daily_stock_in_pieces</th>\n",
       "      <th>daily_sales_in_pieces</th>\n",
       "      <th>daily_sales_in_euros</th>\n",
       "      <th>date_of_day</th>\n",
       "      <th>weekday</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>432969</td>\n",
       "      <td>ACM OBST</td>\n",
       "      <td>Steinobst</td>\n",
       "      <td>Aprikosen</td>\n",
       "      <td>APRIKOSEN 750g</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020-01-17</td>\n",
       "      <td>Friday</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "       purchasing_area_description product_group_category_description  \\\n",
       "432969                    ACM OBST                          Steinobst   \n",
       "\n",
       "       product_subgroup_category_description    article_name  \\\n",
       "432969                             Aprikosen  APRIKOSEN 750g   \n",
       "\n",
       "        daily_shrinkage_in_euros  daily_shrinkage_in_pieces  \\\n",
       "432969                       NaN                        NaN   \n",
       "\n",
       "        daily_stock_in_euros  daily_stock_in_pieces  daily_sales_in_pieces  \\\n",
       "432969                   0.0                    0.0                    NaN   \n",
       "\n",
       "        daily_sales_in_euros date_of_day weekday  \n",
       "432969                   NaN  2020-01-17  Friday  "
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.sample(1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, when our data looks cleaner, we have to explore our rows on missing values. And decide which method we can use to work with them."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "purchasing_area_description                   0\n",
       "product_group_category_description            0\n",
       "product_subgroup_category_description         0\n",
       "article_name                                  0\n",
       "daily_shrinkage_in_euros                 738074\n",
       "daily_shrinkage_in_pieces                738074\n",
       "daily_stock_in_euros                          0\n",
       "daily_stock_in_pieces                         0\n",
       "daily_sales_in_pieces                    724596\n",
       "daily_sales_in_euros                     724596\n",
       "date_of_day                                   0\n",
       "weekday                                       0\n",
       "dtype: int64"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Target columns are \"daily_shrinkage_in_euros\" and \"daily_sales_in_euros\". There are 738074 null values (99%) for \"daily_shrinkage_in_euros\" and 724596 (98%) for daily_sales_in_euros. The main question here is how we can work with null values?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(597, 12)"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df[\"daily_shrinkage_in_euros\"].isnull()== False].shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(14075, 12)"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[df[\"daily_sales_in_euros\"].isnull()== False].shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- There are 597 products that have values for shrinkage and 14075 products that have values for sales. We cannot follow the technique of dropping all null values. \n",
    "- Also we considered a method to estimate some of the missing values by exploring more deeply each product. For this, we checked a few products out of each purchasing area and did not find any pattern which we could use to estimate the missing values. Most of the products do not have a stable day by day record. We did not include massive exploration of each product into the following report, but in general we can use the following function to explore missing values in depth by each article name. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In order to check if we have products with stable data, we used function bellow. It takes time to run, so I embed code for illustration instead of execution in Python:\n",
    "```python\n",
    "gdf = df.groupby([\"article_name\"])\n",
    "def countna(x):\n",
    "    return (x.isna()).sum()\n",
    "gdf.agg(['count', countna, 'size'])\n",
    "```\n",
    "Function above has just approved that for most products there are missing values every day and that's why it is impossible to use any approach for estimation.\n",
    "As a result, to drop data with missing values is not a good choice, because we lose the information and would have ended up with a very small dataset. \n",
    "\n",
    "We made the following decision to check how many columns have both \"NaN\" values for sales and for shrinkage and to drop the rows only in this case. After dropping NaN values, there are 14406 rows."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(14406, 12)"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df.dropna(subset=[\"daily_shrinkage_in_euros\", \"daily_sales_in_euros\"], how=\"all\")\n",
    "df.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We had a few consulting sessions with data analysts from \"Metro\". The main insights are as following:\n",
    "- most of the data for shrinkage has been done by employees and has a human factors( wrong inputs);\n",
    "- the reasons behind NaN values for \"daily_shrinkage\" and \"daily_sales\" is missing information which could be filled with zero values and interpreted as \"not exist\"."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "df[\"daily_sales_in_pieces\"] = df[\"daily_sales_in_pieces\"].fillna(0)\n",
    "df[\"daily_sales_in_euros\"] = df[\"daily_sales_in_euros\"].fillna(0)\n",
    "df[\"daily_shrinkage_in_pieces\"] = df[\"daily_shrinkage_in_pieces\"].fillna(0)\n",
    "df[\"daily_shrinkage_in_euros\"] = df[\"daily_shrinkage_in_euros\"].fillna(0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In addition, we discussed business logic behind shrinkage and came to the conclusion, that:\n",
    "- shinkage could be only negative. So positive shrinkage is a mistake of the system or wrong input. \n",
    "- sales could be only positive and negative values is a mistake.\n",
    "\n",
    "Let's determine what percentage of our data has invalid values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0005553241704845203"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df[df[\"daily_shrinkage_in_euros\"] > 0])/df.shape[0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.006455643481882549"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(df[df[\"daily_sales_in_euros\"] < 0])/df.shape[0]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We have explored only 8 rows with positive shrinkage and 93 rows with negative sales. Given that this is less then 1% of data, we will remove these rows."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "purchasing_area_description              14305\n",
       "product_group_category_description       14305\n",
       "product_subgroup_category_description    14305\n",
       "article_name                             14305\n",
       "daily_shrinkage_in_euros                 14305\n",
       "daily_shrinkage_in_pieces                14305\n",
       "daily_stock_in_euros                     14305\n",
       "daily_stock_in_pieces                    14305\n",
       "daily_sales_in_pieces                    14305\n",
       "daily_sales_in_euros                     14305\n",
       "date_of_day                              14305\n",
       "weekday                                  14305\n",
       "dtype: int64"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df = df[df[\"daily_shrinkage_in_euros\"] <= 0]\n",
    "df = df[df[\"daily_sales_in_euros\"] >= 0]\n",
    "df.count()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "During the cleaning process, we have done the following steps:\n",
    "1. changed data types for columns;\n",
    "2. worked with the date column and extracted information about weekdays;\n",
    "3. removed columns with redundant information;\n",
    "4. Worked out rows with the null values by filling missing values for shrinkage in euros and sales in euros with zeroes.\n",
    "5. removed rows with positive values for shrinkage and negative values for sales.\n",
    "\n",
    "We have 14305 rows to analyse. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## iii. Exploring the Data"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Exploring trends of sales and shrinkage\n",
    "\n",
    "We'll begin by exploring our trends of shrinkage and sales by each day during a given month. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "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>daily_shrinkage_in_euros</th>\n",
       "      <th>daily_sales_in_euros</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>date_of_day</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>2020-01-03</td>\n",
       "      <td>-4435.6768</td>\n",
       "      <td>31093.260</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2020-01-08</td>\n",
       "      <td>-839.8867</td>\n",
       "      <td>30303.453</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "             daily_shrinkage_in_euros  daily_sales_in_euros\n",
       "date_of_day                                                \n",
       "2020-01-03                 -4435.6768             31093.260\n",
       "2020-01-08                  -839.8867             30303.453"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_by_date= df.groupby(\"date_of_day\")[\"daily_shrinkage_in_euros\", \"daily_sales_in_euros\"].sum().reset_index()\n",
    "df_by_date=df_by_date.set_index(\"date_of_day\", drop=True, append=False, inplace=False)\n",
    "df_by_date= df_by_date[df_by_date[\"daily_sales_in_euros\"]>0]\n",
    "df_by_date.sample(2)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "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": [
    "fig, ax = plt.subplots()\n",
    "\n",
    "sns.pointplot( x=df_by_date.index,y='daily_shrinkage_in_euros',data=df_by_date,color='lightblue')\n",
    "sns.pointplot( x=df_by_date.index, y='daily_sales_in_euros',data=df_by_date,color='lightgreen')\n",
    "plt.xlabel('Date, YYYY-MM-DD',fontsize = 12, labelpad=20)\n",
    "plt.ylabel('Total amount by day, EUR',fontsize = 12 )\n",
    "\n",
    "ax.set_xticklabels(ax.get_xticklabels(), rotation=90)\n",
    "\n",
    "plt.gca().xaxis.set_major_formatter(plt.FixedFormatter(df_by_date.index.to_series().dt.strftime(\"%Y-%m-%d\")))\n",
    "shrinkage_patch = mpatches.Patch(color='lightblue', label='Shrinkage')\n",
    "sales_patch = mpatches.Patch(color='lightgreen', label='Sales')\n",
    "plt.legend(handles=[shrinkage_patch, sales_patch])\n",
    " \n",
    "plt.grid()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Figure 1. Trends of shrinkage, sales in euros between 1st and 31st of January 2020.* "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Figure 1 shows that shrinkage seems to stay almost the same low during a month. There is some bigger raise of shrinkage on the 2nd of January, it could be affected by holidays, probably we could assume that the store had a bigger amount of products for New Year holidays. Also Figure 1 shows sales drop on specific dates (10th, 17 th, 24th of January), it is Friday so we are guessing that there is some dependency on weekdays."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Exploring weekdays"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "On the previous step we assumed that sales depends on a weekday. We will explore the assumption more by creating  a new dataframe where the main variables are grouped by a weekday. Department store is not getting any input on Saturday and Sunday. "
   ]
  },
  {
   "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>weekday</th>\n",
       "      <th>daily_shrinkage_in_euros</th>\n",
       "      <th>daily_sales_in_euros</th>\n",
       "      <th>absolute_shrinkage</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Monday</td>\n",
       "      <td>-2863.1071</td>\n",
       "      <td>166579.502</td>\n",
       "      <td>2863.1071</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>Tuesday</td>\n",
       "      <td>-4054.3495</td>\n",
       "      <td>141190.966</td>\n",
       "      <td>4054.3495</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Wednesday</td>\n",
       "      <td>-2819.0344</td>\n",
       "      <td>118766.990</td>\n",
       "      <td>2819.0344</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>Thursday</td>\n",
       "      <td>-3477.4796</td>\n",
       "      <td>163614.560</td>\n",
       "      <td>3477.4796</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>Friday</td>\n",
       "      <td>-9617.0413</td>\n",
       "      <td>113097.963</td>\n",
       "      <td>9617.0413</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "     weekday  daily_shrinkage_in_euros  daily_sales_in_euros  \\\n",
       "0     Monday                -2863.1071            166579.502   \n",
       "1    Tuesday                -4054.3495            141190.966   \n",
       "2  Wednesday                -2819.0344            118766.990   \n",
       "3   Thursday                -3477.4796            163614.560   \n",
       "4     Friday                -9617.0413            113097.963   \n",
       "\n",
       "   absolute_shrinkage  \n",
       "0           2863.1071  \n",
       "1           4054.3495  \n",
       "2           2819.0344  \n",
       "3           3477.4796  \n",
       "4           9617.0413  "
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "arrange = ['Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday', 'Sunday']\n",
    "df_by_weekday=df.groupby(\"weekday\")[\"daily_shrinkage_in_euros\",\"daily_sales_in_euros\"].sum().reindex(arrange)\n",
    "df_by_weekday.reset_index(inplace=True)\n",
    "df_by_weekday = df_by_weekday.drop(index=[5, 6])\n",
    "df_by_weekday[\"absolute_shrinkage\"] = df_by_weekday[\"daily_shrinkage_in_euros\"].apply(abs)\n",
    "df_by_weekday"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1a1728bed0>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 360x504 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "ax = df_by_weekday.plot.bar(x=\"weekday\", y= [\"absolute_shrinkage\",\"daily_sales_in_euros\"],color=[\"lightblue\", \"lightgreen\"], rot=0, subplots=False, figsize=(5, 7))\n",
    "plt.ylabel('Total amount, EUR',fontsize = 12,color='black')\n",
    "plt.xlabel('Weekday',fontsize = 12, labelpad=20)\n",
    "shrinkage_patch = mpatches.Patch(color = \"lightblue\", label='Shrinkage')\n",
    "sales_patch = mpatches.Patch(color = \"lightgreen\", label='Sales')\n",
    "plt.legend(handles=[shrinkage_patch, sales_patch])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Figure 2. Total amount of shrinkage, sales in euros by weekday* "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Figure 2 proves that daily sales on Friday are less than other weekdays. The peak of sales takes place on Mondays and on Thursdays. To make a clear vision of sales and shrinkage, we will find a proportion by weekday. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1a17bf4e50>"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "my_explode = (0.1, 0, 0, 0, 0)\n",
    "plt.pie(df_by_weekday[\"daily_sales_in_euros\"], labels=df_by_weekday['weekday'], autopct='%1.1f%%', startangle=15, shadow = True, explode=my_explode)\n",
    "plt.axis('equal')\n",
    "plt.legend(title=\"Weekday\", loc='upper left', labels=df_by_weekday['weekday'], frameon=True, ncol=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Figure 3. Proportion of sales by weekday* "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1a17efdc10>"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "my_explode = (0, 0, 0, 0, 0.1)\n",
    "plt.pie(df_by_weekday[\"absolute_shrinkage\"], labels=df_by_weekday['weekday'], autopct='%1.1f%%', startangle=15, shadow = True, explode=my_explode)\n",
    "plt.axis('equal')\n",
    "plt.legend(title=\"Weekday\", loc='upper left', labels=df_by_weekday['weekday'], frameon=True, ncol=1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Figure 4. Proportion of shrinkage by weekday* "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As a result, Figure 3 and Figure 4 shows that sales takes the biggest part on Mondays and on Thursdays, the least part on Fridays, on contrary shrinkage is reaching the highest value on Fridays (42,7%). "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Exploring freshness"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "scrolled": true
   },
   "outputs": [],
   "source": [
    "df['freshness'] = 'fresh'\n",
    "df['freshness'][df['purchasing_area_description'] == 'ACM FLEISCH'] = 'ultrafresh'\n",
    "df['freshness'][df['purchasing_area_description'] == 'ACM FRISCHFISCH'] = 'ultrafresh'\n",
    "df['freshness'][df['purchasing_area_description'] == 'ACM GEMUESE'] = 'ultrafresh'\n",
    "df['freshness'][df['purchasing_area_description'] == 'ACM BLUMEN'] = 'ultrafresh'\n",
    "df['freshness'][df['purchasing_area_description'] == 'ACM OBST'] = 'ultrafresh'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now we have information about each product whether it belongs to the fresh or the ultrafresh category. \n"
   ]
  },
  {
   "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>freshness</th>\n",
       "      <th>daily_shrinkage_in_euros</th>\n",
       "      <th>daily_sales_in_euros</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>fresh</td>\n",
       "      <td>-16797.2265</td>\n",
       "      <td>394066.109</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>ultrafresh</td>\n",
       "      <td>-6034.9998</td>\n",
       "      <td>309183.872</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "    freshness  daily_shrinkage_in_euros  daily_sales_in_euros\n",
       "0       fresh               -16797.2265            394066.109\n",
       "1  ultrafresh                -6034.9998            309183.872"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_freshness = df.groupby(\"freshness\")[\"daily_shrinkage_in_euros\", \"daily_sales_in_euros\" ].sum().reset_index()\n",
    "df_freshness"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's find proportions of fresh and ultra fresh products in the shrinkage (absolute values) and sales."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.legend.Legend at 0x1a17ff0350>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x360 with 2 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_pie = pd.DataFrame({'fresh':[16797.2265,394066.109],\n",
    "               'ultrafresh':[6034.9998, 309183.872],\n",
    "               }, index=['daily_shrinkage_in_euros','daily_sales_in_euros' ])    \n",
    "title= ['Daily shrinkage, %','Daily sales, %' ]\n",
    "fig, axs = plt.subplots(nrows=1, ncols=df_pie.index.size, figsize=(15,5))\n",
    "colors = [\"lightblue\",\"lightgreen\"]\n",
    "fig.subplots_adjust(hspace=1.5)\n",
    "\n",
    "\n",
    "for row in range(df_pie.index.size):\n",
    "    fig.add_subplot(axs[row] )\n",
    "    plt.pie(df_pie.loc[df_pie.index[row],:], labels=df_pie.columns, autopct=\"%1.1f%%\", colors=colors)\n",
    "    axs[row].set_title(title[row])\n",
    "fig.legend(title=\"Freshness\", loc='upper right', labels=[\"Fresh\", \"Ultrafresh\"], ncol=2)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Figure 5. Proportion of Fresh and Ultrafresh products in shrinkage and sales* "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Figure 5 demonstrates that fresh products take the main part in shrinkage and sales. So following the data for only one month we can conclude that the current depo has less ultrafresh product then normal fresh.  We will also check how many unique products each group has and find out a proportion. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "freshness\n",
       "fresh         1026\n",
       "ultrafresh     716\n",
       "Name: article_name, dtype: int64"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby(\"freshness\")[\"article_name\"].nunique()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "freshness\n",
       "fresh         58.897819\n",
       "ultrafresh    41.102181\n",
       "Name: article_name, dtype: float64"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby(\"freshness\")[\"article_name\"].nunique()/df[\"article_name\"].nunique() * 100"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "58,9% of unique products belong to normal fresh and 41,1% belong to the ultrafresh (almost equal parts).\n",
    "\n",
    "On this step, we found out that shrinkage for January 2020 contains a bigger part of the normal fresh products, then the ultrafresh products."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Exploring the top best-selling products "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "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": [
    "df_by_name_top_sale = df.groupby(\"article_name\")[\"daily_sales_in_euros\",\"daily_shrinkage_in_euros\"].sum().reset_index()\n",
    "df_by_name_top_sale[\"% out of total sales\"]= (df_by_name_top_sale[\"daily_sales_in_euros\"]/df_by_name_top_sale[\"daily_sales_in_euros\"].sum())* 100\n",
    "\n",
    "df_by_name_top_sale = df_by_name_top_sale.nlargest(10, [\"daily_sales_in_euros\"])\n",
    "df_by_name_top_sale.set_index(\"article_name\", inplace=True, drop=\"true\")\n",
    "\n",
    "ax = df_by_name_top_sale[\"% out of total sales\"][:10].plot.barh(color=\"lightgreen\")\n",
    "plt.xlabel(\"Proportion of sales for the top 10 products, %\", fontsize = 12, labelpad=20);\n",
    "plt.ylabel(\"Article name\", fontsize = 12,);\n",
    "\n",
    "for patch in ax.patches:\n",
    "    ax.text(\n",
    "        patch.get_width() + 0.1, \n",
    "        patch.get_y() + 0.38,\n",
    "        \" {} %\".format(patch.get_width().round(2)), \n",
    "        fontsize=10,\n",
    "        color='black'\n",
    "    )\n",
    "\n",
    "ax.invert_yaxis()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Figure 6. Top 10 best-selling Products in the department store for January 2020*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Based on the sales of products across different categories in the department store, the top 10 best selling products are shown on Figure 6:\n",
    "\n",
    "1. 10er Eier b/w L-M BIO\n",
    "2. MC AVOCADO \n",
    "3. 1L ARO H-SCHLAGSAHNE\n",
    "\n",
    "....\n",
    "\n",
    "Top 10 best-selling products in the department store for January 2020 make almost 11% of total sales. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAlAAAAEZCAYAAABYXRt8AAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nOzdeZhUxdn38e8PHFBEBTQacFBUiCjbgKgYXCCKIApEBQUxLnF58mBeXIJIBJSIBoNiMEbjQoyIIIobSAQkGFSMOGzDpiKiRBEfBRJIYAYY4H7/ONV46Onu6cEZ1vtzXX1Nd22n6vQwfVNV57TMDOecc845l71Ku7sDzjnnnHN7Gw+gnHPOOefKyAMo55xzzrky8gDKOeecc66MPIByzjnnnCujA3Z3B5zbXx1xxBFWr1693d0N55zbq8yZM2e1mf1gd/fDAyjndpN69eoxe/bs3d0N55zbq0j65+7uA/gSnnPOOedcmXkA5ZxzzjlXRh5AOeecc86VkQdQzjnnnHNl5JvIndtN1m4s5pUlX+/ubjjn3C51yYm1d3cXyoXPQDnnnHNuj1dYWMiFF14I0EjSYkn3pyonKUfSSEkLJX0k6dcpylSVNFnSIkm9YulPSmqeTX88gCqFpOXhTSiQNDuWPl1Sy51s8xpJq0KbBZKeDenPSPo8pM2XdG6q40n6eejTgvDmd4mV6yPp45A+X9JVqforqZ6kReF5G0nrYv0pkHReyNsaXi+S9LqkGmnGtD72vKOkpZKOkTRIUp8U5XMljQ/llkl6WFKVkFdN0ugwxkWSZkiqHvL6h384C0K/Tpf0anj+adI4fhzGvSSW9lJoZ4d+SbotnLeF4bw9JCknRb/j7X0k6caQPkbS/8bKnR766LO8zjlXTvr06QOwGGgOtJZ0QYpi3YCqZtYEOAX4H0n1ksq0B+YATYHE3/FmQCUzm5dNX/yPe3bamtnqcm7zBTP7ZYr0283sJUltgSeBBvFMSblAf6CFma0LgcUPQt4vgHbAaWb2H0mHAT/Nsj/vmtlFKdKLzCwvtD8SuAm4L10jIeh7BDjfzL6QlKqMgFeAP5lZF0mVw1jvA24Hbga+Cb/8SDoRKJZ0BnBRGPsmSUcAVczs4lCuDdAnPo5w/J5mlvaGS+G8nQ+0MrO1IZC7DTgIKE5RpaeZzZZUC1gm6RngVuD9EKCtAf4I9DKzLemO65xzLnvVqlWjbdu2AJjZZklzgdwURQ04OPwH9iBgM/CfpDLFIS8eBw0GfpFtf3wG6nuSVEnRVOG94fV1kj4JMxVPSfrjTjb9PnB0ivQjgf8C6wHMbL2ZfR7y7iT60P5PyFtnZiN38vhl6RMAks4CngIuNLNlGdr5CbDRzP4S+rmVKAD5uaRqQG3gq0RhM1tiZptC+urwHDNbbWYrv+eYIApI/9fM1oZ2N5vZ/YnzmEF1YAOw1cy+AR4EhhL9A1xgZjPKoW/OOeeShNWQTsC0FNkvEf1t/hr4AnjQzP6VVGYq8EPgA2CopM7AnLJ8pvgMVOkMeFOSAU+Y2ZOxvAOA0cAiM7tPUh1gINCCKMh5C5ifpt3LJZ0Znj+cCCZiOgCvpag3H/gG+FzSNOAVM3td0iHAIaUELqMlFYXnVYBtsbyzJBXEXl8abyvMEp0L/DlN21WB8UAbM/s4Qx8AGhFNnW4XZsy+AOoDTxOd865E/zhGmtlS4E3gLkmfAH8jmsV7u5RjwY7jnmpmt8fGdQhQPRaEZmO0pE1Es4O3hAAQ4HHgaqANkHJ5Nyz53QhwRJ20sahzzrnMngf+YGafpcg7DdgK1AFqAu9K+lu8bFgduAKiPVPAFKCzpIeAY4BnzWxCpg74DFTpWptZC+AC4CZJZ8fyniAET+H1acDbZvYvMysGxmVo9wUzywuPePD0gKTPgOeA3yZXCh/WHYCuwCfA7yUNAkQU7GXSM3FMoGNS3rux/uTFgqeDQmC1BqhFFLWnUgz8A7iulD6Qoa8CzMwKgOOBB8IxZ0k6yczWE61n3wisAl6QdE0Wx+sZG9ftSXk79EVS+7C/abmkH2dorynRP7I+ko4l6vg2ot+JSWa2JlVFM3vSzFqaWcvDah6eRdedc84lqQcsNbPhafKvACabWbGZfQu8R5r/1Aa9gJHAGUTLfZcDA0rrhAdQpUhM54U34VWiICnhH0BbSQeG1yU3/JTd7USzMAOI3tBUfTIzyzezIUB3otmi/wAbJB1fDn2IS+yBOpZo1uqmNOW2AZcBp0q6s5Q2F5P0yyzpUKAusAy2L02+Yma9iILJjiF9q5lNN7O7gV8Cl+7csCKx83ZceD0ljHcR0Xgz1V0FzAVOjyVvY8eZPeecc+VkwIABAJWBWzIU+wL4iSIHA62AlCsjkmoS7a19FqhG9PfbgANTlY/zACoDSQeHJR7Cm3A+0Qdrwp+BN4BxYbNaPnCOpJrh9U59uIeZjIeBSpLaJ/WpjqQWsaQ8IPHFikOAR0MwgqRDE1eJfV9mtg7oTTTjUuLqtFCmkOgXsaekTDNR04Bq+u4KwcrAMOAZMyuU1Dr8UhM2dJ8M/FPSiZLim+rjY/8+hgB/CmvqiU3upf7jCfu1mhOCPueccxVnxYoV3HfffRD9fZ4bVguuB5DUWdI9oeijRHtUFwGzgL+Y2YI0zd4F3GtmRrSM1xJYSLSfNyPfA5XZUcCr4UquA4AxZjY5XsDMHgpXu40CehItu30ArAQ+BNbtzIHNzMLG9L5Eb2pCDvBg2G+1kWgpK3HVwJ+IfmlmSSomWlYbluUhk/dA3WtmLyX1aZ6k+USzXqPS9PtfkjoA70hKXLk4QNItsTK5ki4GHpM0kCiQf4NoEzzACUQBjULeX4GXifaWPRICnS3Ap4T9RKWI74FabWbnJeX/ieh/Hh+EvU3riaZ8013KmmivKlHQNydNOeecc+UkNzcXM0PSYjPbYRUj7FeaEJ6vJ7qVQanM7NbY841EEyVZURR0ufIiqbqZrQ8zUK8CT5vZq7u7X27PU79xMxv68uTSCzrn3D7k+96JXNKc5ABqd/AZqPI3SNFNKA8kumos1ZV0zlHjwJx95isNnHNuf+MBVDkzsxJ33XbOOefcvsU3kTvnnHPOlZEHUM4555xzZeQBlHPOOedcGXkA5ZxzzjlXRh5AOeecc86VkQdQzjnnnHNl5AGUc84551wZeQDlnHPOOVdGHkA555xzzpWR34ncud1k7cZiXlny9e7uhnMuiX/FksuGz0A555xz5aSwsJALL7yQhg0b0qhRI/r165e27JAhQ6hfvz4nnngiU6ZMAWDjxo2cdtppNGvWjEaNGnH33XenrHvHHXfQtGlTrrrqqu1po0aN4uGHHy7fAbm09qsAStLTkr6VtCgpvZakqZKWhp81v+dxtkoqiD36hfQRkk7+Pm3vZH9aSvrDTtb9taSekp6R1LW8+1aGftRLft/SlDNJo2KvD5C0StLEFGXbSFoX3qMFkv4m6cgy9mu5pCPKUsc5t2/r06cPH3/8MfPmzeO9995j0qRJJcp8+OGHjB07lsWLFzN58mR69erF1q1bqVq1Km+99Rbz58+noKCAyZMnM3PmzB3qrlu3jn/84x8sWLCArVu3snDhQoqKinjmmWfo1avXrhrmfm+/CqCAZ4AOKdL7AdPMrAEwLbzOiqRUy6BFZpYXe9wPYGbXm9mHZWi7crZlMzGz2WbWeyernw+8mU3BNOdiV9sANJZ0UHjdDvgqQ/l3w3vUFJgF3FTRHXTO7buqVatG27ZtAahSpQotWrRgxYoVJcqNHz+e7t27U7VqVY477jjq169Pfn4+kqhevToAxcXFFBcXI2mHupUqVWLz5s2YGUVFReTk5PDAAw/Qu3dvcnJyKn6QDtjPAigzewf4V4qsLsDI8Hwk8FMASQeHWatZkuZJ6hLSr5E0TtLrZBlchHrTJbUMz8+X9L6kuaGt6iF9uaS7JM0AusXqVpb0mSI1JG2TdHbIe1dS/Qz9bZOYgZF0TmxmbJ6kQyTVlvROSFsk6axQ9lCgipmtCt04LxzrE0kXpToXkqpLmhbGtTDWhx1mkCT1kTQodl5+Jyk/tH1Wtuc0jUnAheF5D+D50ioo+gt1CPDv8LqWpNfCzNRMSU1D+uGS3gzn7glAsTauDGMokPREeQXAzrm909q1a3n99dc599xzS+R99dVX1K1bd/vr3Nxcvvoq+r/e1q1bycvL48gjj6Rdu3acfvrpO9Q95JBDuPTSS2nevDnHHXcchx12GLNmzaJLly4VOyC3g/0qgMrgKDP7GiD8TCzj9AfeMrNTgbbAA5IODnlnAFeb2U9StHdQ0hLe5fHMsOQzADjPzFoAs4HbYkU2mtmZZjY2kWBmW4FPgJOBM4E5wFmSqgK5ZvZpKf1N6APcZGZ5wFlAEXAFMCWkNQMKQtnziGbkEuoB5xAFJ49LOjDFudgIXBzG1RYYpuT/PqV2gJmdBtwCpF70z95YoHvoX1Pggwxlz5JUAHxBNN6nQ/pvgHlhZupO4NmQfjcww8yaAxOAYwAknQRcDrQO53Er0DP5YJJulDRb0ux1/17zPYfpnNtTbdmyhR49etC7d2+OP/74EvlmViIt8aeycuXKFBQUsGLFCvLz81m0qOTuhb59+1JQUMCwYcMYOHAg99xzDyNGjOCyyy7j3nvvLf8BuRI8gMrsfKBf+ICdDhxI+MAEpppZqtksKLmE90JSfiuiQOi90PbVwLGx/OTyCe8CZ4fHEKJA6lSipafS+pvwHvCQpN5ADTPbEupfG2aEmpjZf0PZDkSzOQkvmtk2M1sKfAY0THEuBPxW0gLgb8DRwFFpxhP3Svg5hyhQ22lmtiC00QN4o5TiiSW8usBfgKEh/UxgVGjvLeBwSYcRnfvnQvpfCTNWwLnAKcCscP7PBUr81TSzJ82spZm1PKzm4Ts/SOfcHu3GG2+kQYMG3HLLLSnzc3Nz+fLLL7e/XrFiBXXq1NmhTI0aNWjTpg2TJ09Oe5x58+YB8KMf/Yhnn32WF198kUWLFrF06dJyGIXLxAOoyDeSagOEn9+GdAGXxgKhY8zso5C34XscT0RBR6Ldk83sulh+urbfJZo1Oo0oMKgBtAHeyaK/AIT9WNcDBwEzJTUMS5tnE+0VGiUpcVnHaUB+vHpSfxKv4/3tCfwAOCXMxHxDFMhtYcfftwPZ0abwcytZ3l5DUt3YLN8vkrInAA+SxfJdUp2zE82nyLeknzt0BxgZO/cnmtmgMhzbObePGDBgAOvWrWP48OFpy3Tu3JmxY8eyadMmPv/8c5YuXcppp53GqlWrWLt2LQBFRUX87W9/o2HDhmnbScw+FRcXs3XrViDaI1VYWFi+g3IleAAVmUA0C0T4OT48nwL8v8QSlKTm5XS8mUBrSfVDu9Uk/SiLeh8APwa2mdlGoqW2/yEKrLLqr6QTzGyhmf2OaOmwoaRjgW/N7Cngz0ALSY2Aj8PSYUI3SZUknUA0u7IkRR8PC20VS2rLdzNr3wBHhj1EVYGLshhvRmb2ZSxgeTwp+2ngHjNbWIYmzwSWhefvEJbgJLUBVpvZf5LSLwASV2xOA7oqXMUX9lDFZxWdc/uBFStWcN999/Hhhx/SokUL8vLyGDFiBAATJkzgrrvuAqBRo0ZcdtllnHzyyXTo0IFHH32UypUr8/XXX9O2bVuaNm3KqaeeSrt27bjootR/Ll977TVOPfVU6tSpQ40aNTjjjDNo0qQJkmjWrNkuG/P+SqnWYfdVkp4nmrE5gugD/W4z+7Okw4EXiZa7vgC6mdm/FF3JNZwoaBGw3MwuknQN0NLMfpnmOFuB+Af3ZDPrJ2k60MfMZkv6CfA7oGooM8DMJkhaHtpenabtd4mWne6UdAXwGFDLzLZl6G+bcNyLJD1CtDdpK/AhcA3QHbgdKAbWA1cBlxIFDc+E4z5DtFzVkmhJ7jYzm5h8LsL+rteBHKIArzVwgZktD8uGvYHPiWa7lpvZoKTzcgQw28zqJY27HrA0vG8Jt5rZuKRy682selLa9vGnSB8f+iNgHXC9mX0iqRbRkt5xQCFwo5ktCL8rzxP9Dr0NXEI027Y67HX7NdF/TIqJ9prteP1xTP3GzWzoy+mn5p1zu4ffSHPPJmmOmbXc7f3YnwIolz1JU4GrEpvrXflr2bKlzZ49e3d3wznn9ip7SgC1J9y3x+2BzKzd7u6Dc845t6fyPVDOOeecc2XkAZRzzjnnXBl5AOWcc845V0YeQDnnnHPOlZEHUM4555xzZeQBlHPOOedcGXkA5ZxzzjlXRh5AOeecc86VkQdQzjnnnHNl5Hcid243WbuxmFeW+DflOFcR/PvsXEXzGSjnnHMupn///tStW5fq1aunLVNcXMzVV19NkyZNOOmkkxgyZMj2vIcffpjGjRvTqFEjhg8fnrL+I488QuPGjenYsSObN28GYMaMGdx2223lOxhXYTyAcs4552I6depEfn5+xjLjxo1j06ZNLFy4kDlz5vDEE0+wfPlyFi1axFNPPUV+fj7z589n4sSJLF26tET9ESNGsGDBApo3b86UKVMwMwYPHszAgQMraliunO1XAZSk5ZIWSiqQNDuWXkvSVElLw8+aIV2S/iDpU0kLJLXYiWMeIGm1pCFJ6dMlLZE0X9IsSXmxvMMkPStpWXg8K+mwNO1vDeNZJOl1STWS8m+VtDFeX1IbSetCvcTjPEn1JC1Kqj9IUp/Y6z6SPg7Hmy/pqth4WsbKbW8r3fEy9T9VX7I4pzmS7g/v4yJJ+ZIuCHnLJR2RdA4mhudHSZoYxvOhpDfSHPcgSW9LqiwpT9L7khaH343LY+XGSmqQqg3n3J6vVatW1K6deQlQEhs2bGDLli0UFRVRpUoVDj30UD766CNatWpFtWrVOOCAAzjnnHN49dVXU7ZRXFxMYWEhOTk5jBo1io4dO1KzZs2KGJKrAPtVABW0NbM8M2sZS+sHTDOzBsC08BrgAqBBeNwI/Gknjnc+sAS4TJKS8nqaWTPgMeCBWPqfgc/M7AQzOwH4HBiRpv2iMJ7GwL+Am5LyewCzgIuT0t8N9RKPv5U2EEm/ANoBp4XjnQ0kjymddMcrrf+ppDung4HaQOPQXifgkCzauweYambNzOxkvnv/k/0ceMXMtgKFwFVm1gjoAAyPBa9/AvpmcVzn3F6qa9euHHzwwdSuXZtjjjmGPn36UKtWLRo3bsw777zDmjVrKCws5I033uDLL78sUb9Pnz60atWKVatW0bp1a0aOHEmvXr12w0jcztofA6hUugAjw/ORwE9j6c9aZCZQQ1JtSZUkPRZmHyZKekNS1zRt9wAeBr4AWqUp8z5wNICk+sApRMFAwj1AS0knlDKO7e2Etk4AqgMDQj++rzuBXmb2HwAzW2dmI0upUxY79D+DEudUUjXgBuD/mdmm0L9vzOzFLNqrDaxIvDCzBWnK9QTGhzKfmNnS8Hwl8C3wg1DuXeA8SX6RhnP7qPz8fCpXrszKlSv5/PPPGTZsGJ999hknnXQSd9xxB+3ataNDhw40a9aMAw4o+afgZz/7GfPmzeO5557joYceonfv3kyaNImuXbty6623sm3btt0wKlcW+1sAZcCbkuZIujGWfpSZfQ0Qfh4Z0o8G4v91WBHSLgHqAU2A64EzUh1M0kHAucBE4HnSBzEdgNfC85OBgjDLQejTVqAAaJRuYJIqh2NNiCX3CMd9FzhR0pGxvLOSltQyBmeSDgEOMbNlGYqNTrQHJC+DZTxemv6n6ke6c1of+CIR3KXx91j/4jN6jwJ/lvR3Sf0l1Ulx3CrA8Wa2PEXeaUAVYBmAmW0DPgWapSh7o6TZkmav+/eaTEN1zu3BxowZQ4cOHcjJyeHII4+kdevWzJ4d7Qy57rrrmDt3Lu+88w61atWiQYP0K/orV65k1qxZdOnShXvvvZcXXniBqlWrMm3atF01FLeT9rcAqrWZtSBamrtJ0tmllE+1PGXAmcA4M9tmZv8H/D1N/YuAv5tZIfAycHEIFBJGS1oB3AE8EjumpelLqvSDQkCwBqgFTI3ldQfGhg/0V4BusbzkJbVladpPjDnd8eN6JtoDOiblpTpeaf1PpbRzmknbWP+uTySa2RTgeOApoCEwT9IPkuoeAaxNblBSbWAUcG04zwnfAiUCMTN70sxamlnLw2oenmW3nXN7mmOOOYa33noLM2PDhg3MnDmThg0bAvDtt98C8MUXX/DKK6/Qo0f6BYCBAwcyeHC04FBUVIQkKlWqRGFhYcUPwn0v+1UAFZZaMLNvgVeB00LWN+GDMPGB+G1IXwHUjTWRC6wk+30/PYiWcpYDc4DDgbax/J7AccAYolkQgMVAc0nb35vwvBnwUYpjFIWA4FiiWZCbQp2mRHu3pobjd6f0Zbw1QPIOxlrA6jCzs0HS8aW0UVYp+59BunP6KXBMmCkrMzP7l5mNMbOfEe0ZSw6ui4AD4wmSDgX+CgwIS7xxB4Y6zrm9TN++fcnNzaWwsJDc3FwGDRoEwIQJE7jrrrsAuOmmm1i/fj2NGzfm1FNP5dprr6Vp06YAXHrppZx88sl06tSJRx99NO3G8Hnz5gHQvHlzIJq5atKkCXPnzqVDhw4VPEr3fcmstEmFfYOkg4FKZvbf8HwqcI+ZTZb0ALDGzO6X1A+oZWZ9JV0I/JJoNuV04A9mdpqkbsDVQGeifS8fATea2Uux4x1K9KFeN7EnR9K1wJlmdp2k6UAfM5sdlqWWAeea2UeSXiFaxrsn1LsLaGZml6YY13ozqx6eNyfao3MC0b6p/5jZkFjZz4E2REFbHzO7KEV7s4E7zGyapFrATOACM1smqRfRxuzLzew/YYzdzezJ+HhCO/WAiWbWWFKbDMdL1/+jE/XLcE6Hhvfjf8xscwiGzzWz50LA1dLMVod62/sk6SfATDMrDAFYPtEG8VlJff0SaGBmG8OS3iTgdTMrcaMXSQuB8xNLw6nUb9zMhr48OV22c+578Btp7rskzUm6EGy32J9moI4CZkiaT/QB+VczS3x63Q+0k7SU6Cqz+0P6G8BnRB/aTwGJSyReJpqdWgQ8AXwArEs63iXAW4kP+mA80FlS1XhBMysChgGJ2wVcB/xI0e0TlgE/CmkZmdk8YD7RbFN3olm2uFdDOpTck5TYBH8VMCAsq70F/Ca23PYnouXKWYpuMfA20dVo2Uh3vHT9h2jf1orEA/gfMp/TAcAq4MPQv9fC69KcAsyWtIBoI/uI5OApeJNo+RbgMqJZqmtiY8qD6LYIRDNrfptx55zbR+03M1DlTVJ1M1sv6XCigKx12A/l9lFhhuy2sMyXqdytRLN/f85UrmXLlpbYdOqccy47e8oMlF9mvfMmhvv+VAEGe/C07zOzeeFKvcrxqyRTWEu0sdw559w+ygOonWRmbXZ3H9yuZ2ZPZ1HmL7uiL84553af/WkPlHPOOedcufAAyjnnnHOujDyAcs4555wrIw+gnHPOOefKyAMo55xzzrky8gDKOeecc66MPIByzjnnnCsjD6Ccc84558rIb6Tp3G6ydmMxryzxr8tzrjz5lwi7XaVMM1CSKoVvuHfOOef2Kf3796du3bpUr149bZni4mKuvvpqmjRpwkknncSQIUMAWLJkCXl5edsfhx56KMOHDy9R/5FHHqFx48Z07NiRzZs3AzBjxgxuu+22ihmUqzBZBVCSakgaA2wEPg1pnSXdW5Gdc9+RZJJGxV4fIGmVpImxtAskzZb0kaSPJT2Yop1rQr0CSYslvSSpWsgbJOmrkPehpB6xes9I6hqeT5e0JJQrkPRSmj63iR3n7Vh6h1D/U0n9YumjJS2Q9NtY2kBJXTKcl9MkvRPa+1jSiMR4Qv54Se9nPrvbyx4raU6sz79IU265pCOyadM5t/fo1KkT+fn5GcuMGzeOTZs2sXDhQubMmcMTTzzB8uXLOfHEEykoKKCgoIA5c+ZQrVo1Lr744hL1R4wYwYIFC2jevDlTpkzBzBg8eDADBw6sqGG5CpLtDNTjwDrgWGBzSHsfuLwiOuVS2gA0lnRQeN0O+CqRKakx8EfgSjM7CWgMfJamrRfMLM/MGhG9n/H38fdmlgd0AZ6QlJOmjZ6hjTwz65qcGb5o+TGgczhOt5BeGXgUuAA4Gegh6WRJTQHMrClwlqTDwmznaWY2PlUHJB0FjAPuMLMTgZOAycAhsT60AGpIOi7NOOK+Bn4cxn860E9SnSzqOef2Aa1ataJ27cyLLJLYsGEDW7ZsoaioiCpVqnDooYfuUGbatGmccMIJHHvssSnbKC4uprCwkJycHEaNGkXHjh2pWbNmuY3D7RrZBlDnAr3N7GvAAMxsFXBkRXXMpTQJuDA87wE8H8vrC9xnZh8DmNkWM3ssU2OSDgAOBv6dnGdmS4FCYGf/VV8BvGJmX4T2vg3ppwGfmtlnZrYZGEsUrBUDB0mqBFQBtgL3AHdlOMZNwEgzez8cw8zsJTP7JuRfCrwejtG9tA6b2WYz2xReVqWUfx+SDpI0WdIN4fXAMAs2VdLzkvqUdkzn3N6la9euHHzwwdSuXZtjjjmGPn36UKtWrR3KjB07lh49eqSs36dPH1q1asWqVato3bo1I0eOpFevXrui666cZRtArQN2WLKQdAzR/9jdrjMW6C7pQKAp8EEsrzEwJ8t2LpdUQDSDVYsoyNiBpBbA0ljgk2x0bAnvgRT5PwJqhuW+OZKuCulHA1/Gyq0Ajjazj4AvgLnAi0B9QGY2L8M4ShtzIsh8PjwvlaS6khaEPv7OzFamKVqd6LyNMbOnJLUkCtiaA5cALdO0f2NYZp297t9rsumSc24Pkp+fT+XKlVm5ciWff/45w4YN47PPvpvs37x5MxMmTKBbt24p6//sZz9j3rx5PPfcczz00EP07t2bSZMm0bVrV2699Va2bdu2q4bivqdsA6gRwMuS2gKVJJ0BjCRa2nO7iJktAOoRBQNvfI+mXgjLVD8EFgK3x/JulbSEKDgblKGN+BLe7SnyDwBOIZoxaw8MlPQjQCnKJmY1bwntDQMGA3dJ6i/pxcQsT7bC8l59YIaZfQJsCcucGZnZl2EZsT5wdWgnlfHAX8zs2fD6TGC8mRWZ2X9JEZSG9p80s5Zm1vKwmoeXZUjOuUzW/XIAACAASURBVD3AmDFj6NChAzk5ORx55JG0bt2a2bNnb8+fNGkSLVq04Kij0v3piKxcuZJZs2bRpUsX7r33Xl544QWqVq3KtGnTKnoIrpxkG0D9jmhW4FEgB3ia6APk4Qrql0tvAvAgOy7fASwmCliyZmZG9EF/diz592E/0eXAs2G2a2esACab2QYzWw28AzQL6XVj5XKBHWZ5wqbx2UTLi43N7DLgZ/HN4UGmMV9OtPz4uaTlRIFnqct4CWHmaTFwVpoi7wEXSEoEhKkCQ+fcPuaYY47hrbfewszYsGEDM2fOpGHDhtvzn3/++bTLd3EDBw5k8ODBABQVFSGJSpUqUVhYWGF9d+UrqwAq7C0ZbmYnm9nBZnZSeG0V3UFXwtPAPWa2MCn9AeDOMMuTuOVENtfFngksS040s1eIgpird7Kf44k2gx8QAp/TgY+AWUADScdJqkIU1ExIVAqb1m8O46lGmJ0i+l2tknSMPxLNEp0eq3+lpB8SzdJ1MLN6ZlaPKNDKGEBJyk1s0pdUE2gNLElT/C5gDdFGeYAZQCdJB0qqznd71Zxze4m+ffuSm5tLYWEhubm5DBo0CIAJEyZw113RdsybbrqJ9evX07hxY0499VSuvfZamjZtCkBhYSFTp07lkksuyXicefOinQnNmzcH4LrrrqNJkybMnTuXDh06VNDoXHlTtjGQpHpE+252uEGGmY0p9165EiStN7PqSWltgD5mdlF4fRHwG74LPP6avLwm6Rqi4OQroqBkBXCNmX0raRCw3sweDGVPAcYQXd32NDDRzF6SNB2oDRSFZleb2Xkp+nw7cC2wDRhhZsNDekdgOFAZeNrM7ovVuQX4t5mNDLM7Y4j2Or1hZnekOMYZwFCiCxq2Ec10PQxMBXLjQb6kucD/AkcBLc3srqS22gHDwrkT8EczezLFMZcT7XFaE87LKjPrG85fD+CfwCpgupk9lVw/oX7jZjb05cnpsp1zO8FvpLnvkzTHzFLuM92l/cgmgJL0a6L/cS/muw9NiCanzk5dy7n9i6TqZrY+zLi9A9xoZnPTlfcAyrny5wHUvm9PCaCy/SqXXwGnmNmHFdkZ5/ZyT0o6GTiQ6PYKaYMngBoH5vgfe+ec20tlG0CtAZZXYD+c2+uZ2RW7uw/OOed2jWwDqFuI/nc9HNjhvkCJGyU655xzzu0vsg2gqgDnE91dOs6INgI755xzzu03sr0P1GPAncChRPeBSjySLyt3zjnnnNvnZTsDdQDRXZe3VmRnnHPOOef2BtnOQD1I9M30frdl55xzzu33sp2B6k30vWl3StrhG1DN7Jhy75Vzzjnn3B4s2wDqygrthXPOOefcXiSrAMrM3q7ojjjnnHPO7S2ynYFCUh7RN9MfQeyb55O/T8w555xzbl+XVQAl6Ubg98CbwAXAJKL7Qo2vuK45t29bu7GYV5Z8vbu74dwewb/WyO1tsr0Kry/QwcwuBorCz65AcYX1zDnnnMtC//79qVu3LtWrV09bZvTo0eTl5W1/VKpUiYKCAgoLC7nwwgtp2LAhjRo1ol+/finrv/zyyzRq1IizzjqLNWuia6mWLVtG9+7dK2RMbs+XbQB1pJm9G55vk1TJzCYBnSqiU5KWS1ooqUDS7Fh6LUlTJS0NP2uGdEn6g6RPJS2Q1KKMx3tGUtektPXhZ6XQ9qLQp1mSjkvRxnRJS8LxP5b0R0k1Yvlbw3gSj36xemm/VVpSPUlFoc6Hkh4PfWojaWKqcUh6NZT/VNK62DF/HOtnIu2lUHeQpEJJRyafg6T+z5c0V9KP0/TvWUk5Ie8aSX9McZ5ahuc/D+d0QTi/XSQ9GmurKNbPHd6fUP/KUHdx6NeIxDlPN84y1Jsf3uu8WL3lko5IPjex/EGS+qR7L51zFaNTp07k5+dnLNOzZ08KCgooKChg1KhR1KtXj7y86J93nz59+Pjjj5k3bx7vvfcekyZNKlF/2LBhzJw5k6uuuooxY8YAMGDAAAYPHlz+A3J7hWz3QK2QVM/MlgOfAF0krQY2V1jPoK2ZrU5K6wdMM7P7QwDSD7iDaFmxQXicDvwp/CwPlwN1gKZmtk1SLrAhTdmeZjZbUhVgCNES5zkhr8jM8tLUK80yM8uTdADwFvBT4F/pCocZQiS1AfqY2UWJPEW38uppZrNTVF0N/IronCbb3n9J7YnGlxhbon+VganAZcDoTAMK57E/0MLM1kmqDvzAzMaH/HrAxHTnTFIH4FbgAjP7Khz7auAoYG0oVmKcZakn6VrgAaBdprE453avVq1alan8888/T48ePQCoVq0abdu2BaBKlSq0aNGCFStWlKhTqVIlNm3aRGFhIVWrVuXdd9+ldu3aNGjQ4PsPwO2Vsg2ghgInAcuBe4CXiL7GpXfFdCutLkCb8HwkMJ3ow74L8KyZGTBTUg1JtYFvgD8SfdB/TjTj9rSZvUT2agNfm9k2ADMr+S8riZltltQX+FRSMzObX4bjZWp3i6R/APWBzP/d2jlPA9dI+p2ZpQ3QiL7S598p+rdVUj5wdBbHOhL4L7A+1F2feJ6l/kTB4VeJY4f+l2e994Hby9An59xe4IUXXmD8+JJbeNeuXcvrr7/OzTffXCLv7rvvpn379tSpU4fnnnuOyy67jLFjx+6K7ro9VFZLeGb2TFiyI/ysCdQ0sz9VUL8MeFPSHEUb2BOOMrOvQz++JvoQhugD+8tYuRUh7RKgHtAEuB44I8MxH4gvscXSXwQ6hfRhkppnNYDog3k+0DAkHZS0hHd5Nu3ESaoGnAssLGvdJKNj/Xgglr6eKJgo+dfju/5/DIwASsxbSzqQaOZvchZ9mE8U4H4u6S+Syroc3AiYW0qZVOPMpl5CB+C1MvYrI0k3Spotafa6f68pvYJzrlx98MEHVKtWjcaNG++QvmXLFnr06EHv3r05/vjjS9Rr164dc+bM4fXXX+e1116jY8eOLFmyhK5du3LDDTdQWFi4q4bg9hBZ38YAIOyPqR5/bWaflXuvoLWZrQzHmyrpYzN7J1PXUqQZcCYwLswe/Z+kv2do4/b4zFRij4uZrZB0IvCT8JgmqZuZTctiHPF+fZ8lvBNCUGfAeDObJOmcNGUti/bSLeEB/AEokDQsKT2+hHcG8KykxF+gRP8aAC+Z2YJS+mJhtqoDcCpRUPh7SaeY2aAs+r8DSU2AUcAhwJ1m9kLIyjTOTPVGSzoYqAyUaT9daczsSeBJgPqNm2XzXjnnytHYsWO3L9/F3XjjjTRo0IBbbrklY/3CwkJGjhzJlClTOP/88xk/fjxjxoxh9OjR3HDDDRXVbbcHymoGSlIHSV8BXwOfxh5LK6JTZrYy/PwWeBU4LWR9E5bmCD+/DekrgLqxJnKBlaQOrHamP5vMbJKZ3Q78lmgPUkZhb00T4KNy6MIyM8szs+axAGMN0UxgXC2ifUw7zczWAmOAXhnKvE90P7AfxPtHtLTYSlLnbPpokXwzGwJ0By4tQ1cXE4IbM1sYjj8JOKgc6vUEjiM6D4+WoU/OuT3Ytm3bGDduXIkr5wYMGMC6desYPnx4qW0MHTqUm2++mZycHIqKipBEpUqVfAZqP5TtVXiPEi3ZVDezSrFH5fLukKSDJR2SeE50v6lFIXsC0YZfws/xsfSrFGkFrAtLfDOASxVdtXYU3+2fKkt/WkiqE55XApoC/yylTg7RJusvY7Mx5W0pUEfSSeGYxwLNgIKMtbLzEPA/pJmhlNSQaHYm+XsRvyba2P/rkDQLaC3ph6FeS6Aq8KWkOtrxask8SjmvSYYAD4bN6AmlBU9Z1zOzYmAAUUB4Uhn65Zzbxfr27Utubi6FhYXk5uYyaNAgACZMmMBdd313r+d33nmH3NzcHZboVqxYwX333ceHH35IixYtyMvLY8SIESmPs3LlSmbPnk2XLl0A+NWvfkWrVq0YOXIkV1xxRcUN0O2Rsl3Cqwk8ETZpV7SjgFfD1WIHAGPMLLGn5n7gRUnXAV8A3UL6G0BHolmxQuDakP4y0fLQIqKrBz8A1pWxP0cCT0mqGl7nE21MT2W0pE1EQcLfiDa3JxyUtLdqspklbjjyV0mJe2q9b2bdKIWZbZJ0JfCXsPeoGLjezLIZ32hJReH5ajM7L6nt1ZJeJbpaLVX/BVwdluGS234NGCTpLDN7V9LNwBsh+FwP9AhXM+YQBTJ1gI3AKuAXWfQ90cc3JP0AmBRm+9YSvc9TMo0zy3qJYxSFpcw+wHVJ2dUkxS8oeCj8HCBp+xqAmeXinKtQQ4cOZejQoSXSO3fuTOfOnbe/btOmDTNnztyhTG5uLtl+tNWpU4eJE7+7e0y3bt3o1q3UP9duH6VsfnHCBtyPzCybq5z2KJKqm9l6SYcTBT+tzez/dne/nKvfuJkNfTmb/fbO7fv8TuQuW5LmmFna+yfuKtnOQLUCeiu699IOwYeZnV3uvSpfExXdJLEKMNiDJ7enqHFgjn9oOOfcXirbAGpEeOx1zKzN7u6Dc8455/YtWQVQZjayojvinHPOObe3yPYqPOecc845F3gA5ZxzzjlXRh5AOeecc86VkQdQzjnnnHNllO1XuVSVdJ+kzyStC2nnS/plxXbPOeecc27Pk+0M1O+BxkTfEZa48+Zi4H8rolPOOeecc3uybO8DdTFQ38w2SNoGYGZfSTq64rrmnHPOObdnynYGajNJwVb4PrE1qYs755xzzu27sp2BGgeMlHQrgKTawHBgbEV1zLl93dqNxbyy5Ovd3Q3ndin/+iK3r8h2BupOYDmwEKgBLAVWAr+pmG4555xzJfXv35+6detSvXr1jOUWLFjAGWecQaNGjWjSpAkbN27kv//9L3l5edsfRxxxBLfcckuJui+//DKNGjXirLPOYs2aaKFl2bJldO/evULG5PZOWQVQZrbZzG4xs+rAUcAhZnarmW2u2O5lT1JdSX+X9JGkxZJujuXVkjRV0tLws2ZIv03Sn2Plekr6a3h+rKRpkhZImi4pN1bu6tDWUklX70RfO0jKl/SxpAJJL0g6JuQ9I+nzkF4g6R8h/RpJq2Lpz5ZyjPMkvRZ7PUTSXyVVCa+PkrRF0nVJ9VZIWhjG/fdwXo+MHff/JH0Ve105lJkQzscySQ9Jyon1Y12s/JSQfq+kW8LzgyS9Jam/pAMkmaTfxfrUT9KAFGO8PpQ9J5bWLaT9tJTzc33sfC6W9KKkg0qpU19SUagzX9J7khpI+mF4z46MlX1SUp9M7Tnnyq5Tp07k5+dnLLNlyxauvPJKHn/8cRYvXsz06dPJycnhkEMOoaCgYPvj2GOP5ZJLLilRf9iwYcycOZOrrrqKMWPGADBgwAAGDx5cIWNye6e0AZSk41M9gEOA42Kv9xRbgF+Z2UlAK+AmSSeHvH7ANDNrAEwLrwH+AJwiqbWkGsC9wP8LeQ8Cz5pZU+AeYAhEwRhwN3A6cBpwdyIgy4akxsAjwNVm1tDM8oDRQL1YsdvNLC88fhxLfyGWflUZjjkIOBW4NBb0Xg68D/RIUeWsMO5/AHea2beJ4xJ9qfQDsdfbgPHAuHB+TwQOJzpnCX+P9bt9Ut+qAq8C/zCz+0JyEXBZONelWZg0hu7A/CzqAYwOfWoUXnfNos6SUKcZMAboZ2b/BwwDhgJIOpXod2N4lv1wzmWpVatW1K6deRnwzTffpGnTpjRr1gyAww8/nMqVK+9QZunSpXz77becddZZJepXqlSJTZs2UVhYSE5ODu+++y61a9emQYMG5TcQt9fLtAfqU6JbFihDGQMqZ8jfZczsa+Dr8Py/kj4CjgY+BLoAbULRkcB04A4z2yKpF/AYkA88bWafhXInA7eG538HErM57YGpZvYvAElTgQ7A82E25w6i5c2lwCYzS75X1h3Ab83so1jfJ3zvE5CGpDuAnwAdzGxjLKsH8EtgnKQfhiAg2fvAjaUc4nxgrZmNAgjn9GZgmaTSlnhzgBeBxWYWn2HaDDwN3EwUrGYyHWgj6QCgGnAMsKiUOjuI1f13WeoBh8bqPAbMDLNhQ4GbzGxLGdtzzpWDTz75BEm0b9+eVatW0b17d/r27btDmeeff57LL78cqeRH3N1330379u2pU6cOzz33HJdddhljx/qWX7ejtDNQZlbJzCqHn+kee0TwlExSPaA58EFIOioEWIlAa/tSi5n9A/gIOI8wgxDMBy4Nzy8GDpF0OFFQ9mWs3ArgaEl1gIFEs1/tgIZputcImFvKEB6ILXmNjqVfHku/tpQ2AM4Gfg5caGaFicRwfmqa2RzgJeCyNPXb813gmE4jYE48wczWEgWRiRnKtrF+94sV/TWwwcx+laLdR4CrJR1ayvG3EQVR5xG9T6X1N66npILQ14OBN7Koc2IYx2dEAehwADPbRnRftNeAhWb2XqrKkm6UNFvS7HX/9otYnasIW7ZsYcaMGYwePZoZM2bw6quvMm3atB3KjB07lh49Uk3AQ7t27ZgzZw6vv/46r732Gh07dmTJkiV07dqVG264gcLCwpT13P4l2zuRH528TCWpZgga9iiSqgMvA7eY2X+yLN+SaDbkB7GsPsA5kuYB5wBfES0TppqRM6Ilm7fN7F9mVkx05WJpxz48fBh/krRfJr6E1zOWHl/C+0tp7RPNguUA5yal9wBeCM/HUnIZ711J3xIFYC+Qmfju5qrp0uNLePfHyrwDnCmpfnLlEISNAW4q5fiJMXQPj7L8N3F0WIb8IfAJcFsWdRJLeMcDfYHHY32eQxSMP5auspk9aWYtzazlYTUPL0NXnXPZys3N5ZxzzuGII46gWrVqdOzYkblzv/t/6/z589myZQunnHJKxnYKCwsZOXIkvXr14te//jVPP/00p5xyCqNHj85Yz+0fsr0K7zUgNyktl2jvyh4jbFx+meiD8ZVY1jeKbr2QuAXDt7G83wDPAfcR3XEdADNbaWaXmFlzoH9IW0c041Q3Vj+XaAYj01Jn3GKgRWhvTfgAfxLIfEnJzvka6Aj8UdLZsfQewPWSlgOvEO0DOy6WfxbRnqyllL6EtpgoAN0u7CerA3xeSt3pwO3AJEk/TJH/ENESYrVS2nmf6JweambLSilbQpg9mkgUMJbFhBR1toWHc243ad++PQsWLKCwsJAtW7bw9ttvc/LJJ2/Pf/7559POPsUNHTqUm2++mZycHIqKipBEpUqVfAbKAdkHUD8ys4XxhPA63TLVLqdoIfvPwEdm9lBS9gQgcbXc1USbnpHUBLgQ+B1REHOspHYh7whJifPza6I9OQBTgPPDDFxNoj1AU4j2UJ0T0g/gu+W/ZEOB/pJOiqWVFiDsNDP7OPTleUlNw8b6ymZ2tJnVM7N6wANEszfxeoXALcDPQ0CUzptATUk9Yft+ooeAp5L2XKXr3wvAw0RB1KFJeauJgvRrSmnDiN6jO0s7XgZnAmUNvnamjnPue+jbty+5ubkUFhaSm5vLoEGDAJgwYQJ33XUXADVr1uS2227j1FNPJS8vjxYtWnDhhRdub+PFF18sNYBauXIls2fPpkuXLgD86le/olWrVowcOZIrrriiYgbn9irZ3khzlaT6ZvZpIiEsu+xJmzhaAz8DFoZ9LRBdQfYGcD/wYtjk/QXQLQRcfwJuTXzQhw3lz0rKI9p0PkSSES013QRgZv+SNBiYFY5xT2xD+W+J9l2tJNq8vi65k2a2MGyyflbSIUTn8At2nOl5QDtetn9aukFLuolos/qIdGXM7ANJ1wOvA+9ScubwZaLN9UOS6q2QNI5ob88QUjCzbeGWAY+Gq/0qEc3mDEzXnxRt/DHMQI0n2pAf9wDQK4s2/poqXdJ9wHvh9yBZT0ltiC6E+CchUJN0MdDEzO5JUefE8PslYBOlb7J3zpWjoUOHMnTo0BLpnTt3pnPnzttfX3nllVx55ZUp2/jss89SpsfVqVOHiRMnbn/drVs3unXrthM9dvsqRf95L6WQdCfRZe/9gc+AE4DBwItm9tsK7eFeRFJ1M1sfZmFeJbqqb49a5nR7jpYtW9rs2bN3dzecc26vImmOmbUsvWTFynYG6n6gmOjeSHWJrkIbQbRU474zSNJ5wIFES1tluSLMOeecc3uJrAKosMn2gfBwaZiZ33naOeec2w+kDaAknW1m74TnP0lXzszeqoiOOeecc87tqTLNQD0GNA7P/5ymjPHdzRKdc8455/YLaQMoM2sce35cunLOOeecc/ubbO9EPj5N+iup0p1zzjnn9mXZ3kizbZr0NuXUD+ecc865vUbGq/AkJW4kWCX2POF4opsPOuecc87tV0q7jUHiO98qseP3vxnRvaAGVUCfnHPOOef2aBkDKDO7Nnwf3D+AZ81s067plnPOOefcnqvUG2mG7zp7yMye2hUdcm5/sXZjMa8s+Xp3d8O5CnXJibV3dxecqxDZbiJ/R1KrCu2Jc845l6R///7UrVuX6tWrZyy3YMECzjjjDBo1akSTJk3YuHEjAB06dKBZs2Y0atSIX/ziF2zdurVE3UceeYTGjRvTsWNHNm/eDMCMGTO47bbbyn9Abp+RbQD1T2CSpGckDZZ0T+JRkZ1zzjm3f+vUqRP5+fkZy2zZsoUrr7ySxx9/nMWLFzN9+nRycnIAePHFF5k/fz6LFi1i1apVjBs3rkT9ESNGsGDBApo3b86UKVMwMwYPHszAgQMrZExu35BtAHUQ0RfjGpBLtKE88Sg3kupK+rukjyQtlnRzLK+WpKmSloafNUO6JP1B0qeSFkhqUYbjVZZUkPRYLemFkD9dUsvwfLmkhbFyf4i1c5ukj0P+fEkPScrJVC8Eo19JqhpeHyFpeXheT1JRKP+hpGcT7YX8MyXlh2N+LOnGWN4gSSW+k0/S+hRpg0If4uOvkVSmUji/i8I4Zkk6TtIHofwXklbF6tdLqj89lFEs7bXk/ki6VdJGSYfF0q6R9McU7SW/JwskvS3p2Fi5rUnj6pdi/M9I+jzkz5V0Ror0+ZLOTTr+kli7L4X0E0NeQfj9fTL5eM65smvVqhW1a2deBnzzzTdp2rQpzZo1A+Dwww+ncuXKABx66KFAFGRt3ryZ2J+iHRQXF1NYWEhOTg6jRo2iY8eO1KxZsxxH4vY12X6Z8LWp0hVtMC9PW4BfmdlcSYcAcyRNNbMPgX7ANDO7P3wY9gPuAC4AGoTH6cCfws9SmdlWIC/xWlJtIB8YnKZKWzNbHU+Q9AvgfKCVma2VVAW4jSjoLE5XL9gK/Dz0OdkyM8uTVBmYClwGjJb0Q2AM8NNwno4Apkj6ysz+ms24k/zezB7MkH85UAdoGvbD5QIbzOx0iIIcoKWZ/TJDG2uB1sCMEKCl+mvYA5gFXAw8U4b+tzWz1ZJ+AwwAbgjpRWaWl6Fewu1m9pKk84EngKZJ6W2BJ4l+vxJ6mtnspHb+QHQuxwNIalKGMTjnvodPPvkESbRv355Vq1bRvXt3+vbtuz2/ffv25Ofnc8EFF9C1a9cS9fv06UOrVq1o1KgRrVu35qc//SmTJ0/elUNwe6GdCoAkNZH0ALCiPDtjZl+b2dzw/L/AR8DRIbsLMDI8Hwn8NJb+rEVmAjUk1Q4zJ48pmsmaKOkNSSX/5Xw3JoV2HzCzRWXodn/gf81sbej3ZjO738z+k0Xd4cCtkjJ9pc5WoqAucR5uAp6JnafVQF+igPL/t3fnYVIVVx/Hvz8QQRwVN4w4RETckUVBMZpXTDRRoxAjUYhrojHGEISImKjAqCQuKK5JjFtwIUqCQFBRY1QEorKPoHEBlSiCiigozCDDeN4/qhovTXdP98DMMMP5PE8/dFfVvVV1bzf3dFXdnpqwO7DEzL6K9S0ys88K3McjQO/4/EfAer9gL2lvoIgQAPWpZjtf4utjVB2TgXYbsd/dSXwezGzeRrTFOVeAtWvXMnXqVEaNGsXUqVMZN24czz777Lr8p59+miVLlvDll1/y3HPPbbD9WWedxZw5c3jooYcYMWIE/fr148knn6RXr14MGDCAr776qja74+qJvAMoSbtKuljSbKAUOAy4uIrNqi1OBXUGpsWk3cxsCYRAC2gZ0/cg/CZVyqKY9iOgDXAwcD5wRBVVDiCMgN2eo8zziambAXGUrMjM3q1i3+ttl0h/D5gKnJVtQ0nNCCNqqa9DBwGz0orNjOnVMSDRtucz5P8dODnm3ySpczXqeBb4vzia1hsYnZbfB3gYmALsJ6klhTueMM2csk3aFN7pVWx/MpAp6EnfL4SRwNR+h8e0m4HnJD0Z3xstyEDSBZJmSpq54rNleXTLOVeV4uJijj76aHbZZReaN2/OiSeeyOzZs9cr06xZM3r06ME//5nxL5MBsHjxYmbMmEHPnj0ZNmwYo0ePpmnTpusFY86l5AygJDWRdKqkx4APgF8A4whTMj82sw1X420CkoqAR4H+eYzkZJrQNuAo4B9m9pWZfQhkCg5S9XUE+gM/NTPLUdcxZtYpPm6Oda8rL+n78aK6UNK3cmyX9AfgUjY8F3tLKgWWAe+Z2dxEfzO1MVe7c7k50bYN/mSPmS0C9gN+B3wFPJtcE5SnSkKgeDqwjZktTMvvDTwSR7nGAj9OVZ9lf8n05yV9DBxLmNpMKU/0q5OZpQdtKcPjcb4AOC8t/R3gIcI5Sjojsd9LAczsr8ABwD8If+LoZcX1bes13OwuM+tiZl122HHnLE1yzhXi+9//PnPnzqWsrIy1a9fywgsvcOCBB7Jy5UqWLAk/FbJ27VomTpzI/vvvn3U/gwcP5pprwgqO8vJyJNGoUSPKyspqpR+ufqlqBOojwrqQNwlrfA40s2uANTXVIIXF0o8Co8wsOdXzUVyjlFqr9HFMX8T6i9mLgcVkDqwy1bcNMAq4yMw+KqStMbhbJWmv+PrpuO7mVWDrPPexgDCid1pa1ttxX+2AbpJ6xPTXgC5pZQ8F/ltI2wthZl+a2ZMxWPgDX0+fFuIRwuje35OJkjoQ1hc9o7CIvjdfT+MtA9JXJg7OlwAAIABJREFUce4EJNeTHQPsSTgu1bkr9NIYCB2XNnV7KeHYX8nXU8c5mdliM7vPzHoSRjPbV6M9zrmEQYMGUVxcTFlZGcXFxZSUlAAwYcIEhgwZAsCOO+7Ib37zG7p27UqnTp045JBD+MEPfsCqVavo0aPHugXmLVu25MILL8xYz5w5cwDo3DkMsp933nkcfPDBzJ49m+OPP77mO+rqnaoWkc8ljOQcDsyX9G411r/kLa5Duhd43cxGpGVPAM4Brov//jOR3lfSI7GdK8xsiaSpwDmS7gd2JYwK/I0N3Qi8YGaPV7PZ1wJ/ltQ7LiIX0KzAffweyLgAPPblt4QRoAnAH4FpksaaWamknYHrqV7wUCWFuxo/NLPF8aaBDoT3RaGmEI7Vw2npfYASM7s2Uee7CnfUzQDukPQNM/tQ4e67pqw/ZYuZlUvqD8yTNMzMPq1G+zYQF83fSngffd/Mns5WVtLxhJscKuJC/50Jo7bOuY1www03cMMNN2yQ3qNHD3r06LHu9ZlnnsmZZ565XpnddtuNGTNm5FVP586duffee9e97t+/P/37969mq92WoKo/5dI9XsjOBgYCt0n6F7At0CTXttV0JGE90Lw4rQJwuZlNJAROf5d0HmHtUGqaZyJwIrAAKANSdww+CnyXMBr0FmEt1YpkZZJaARcBbyTqA3jNzM7I0L7nJaV+hW2umZ1NuIOuOSGo+RJYCfwHmFPFduuY2WtxbVm2n2AYD5RI+raZTZF0JnB3XIMl4BYzeyxR/soYUKT2Xww0l5Rc9J8KUAfE/aX8MG2KrWWsKzUdNR1Y76cF8hGnRjPd7debcCdl0jigt5ldr/BTFhNj8LYS6JNa0J62/yWSHiYssr+GuAYqUeQpMyt4ob2ZmaRhhIX6qQBqlKTy+PwTMzuWcCfmrZJWx/RL49Sxc865Bki5l/ykFZaOIgRTpxGmKO4zs0G5t6o7korMbGUcpZkOHOkXNbe56NKli82cmf5rCM4553KRNMvM0pey1Lq8fgcqxcymEn7Lpx/h93rOrmKTuvZ4vBtqa+AaD56cc845tykUFEClmNlqwlqW9PUsmxUz617XbXDOOedcw7Opf0ncOeecc67B8wDKOeecc65AHkA555xzzhXIAyjnnHPOuQJ5AOWcc845VyAPoJxzzjnnCuQBlHPOOedcgTyAcs4555wrULV+SNM5t/GWr65g7JtL6roZzhXsR/vtXtdNcK7O+QiUc865WnXFFVfQunVrioqKspZZuHAh22yzDZ06daJTp05ceOGFG5Tp0aMH7du3z7j97bffTvv27TnxxBNZs2YNAFOnTuU3v/nNpumE2+J5AFUPSGot6XlJr0t6TdLFibwSSR9IKo2PExN5v5O0QNKbkr5fYJ0lkgbG580kPSNp6KbrVfVJOlfSHVnyVmZJr4zH5xVJsyV9K496+ktqnnh9efVb7ZxLOfnkk5k+fXqV5fbee29KS0spLS3lzjvvXC9v7NixOQOwe+65h7lz59K5c2eefvppzIxrrrmGwYMHb3T7nQMPoOqLtcAlZnYA0A34laQDE/k3m1mn+JgIEPN7AwcBxwN/ktS40IolbQ08Cswys6sK2K5xrtd1oDwen47A74Br89imP9A88brgAGoz6Ldzm51u3bqx++7VnwZcuXIlI0aM4Morr8xZrqKigrKyMpo0acKDDz7IiSeeyI477ljtep1L8gCqHjCzJWY2Oz7/Angd2KOKzXoCj5jZl2b2LrAAOAxA0mBJb8RRpYdTI00ZbAU8Asw3s9+mEiWdKWl6HNH5SypIkLRS0tWSpgFHSFooaYikqcBvJc1O7GMfSbPi8xNje6ZKuk3S4zF9J0njJc2V9LKkDukNlLSXpJckzZB0TR6HE2B74LO4ffdUffH1HXGEqx/QCng+jv5dB2wT+zyqkOOQZ5ucc2neffddOnfuzNFHH82UKVPWpQ8ePJhLLrmE5s2bZ9124MCBdOvWjaVLl3LkkUdy//33c9FFF9VGs90WwgOoekZSG6AzMC2R3DcGGfdJSn292gN4P1FmEbCHpC7AqXEfPwK65KhuELDWzPon6j8AOB040sw6AZXAGTF7W+BVMzvczKbGtNVmdpSZ/R5YIalTTP8pMFJSM+AvwAlmdhSwa6L+q4A5ZtaBMPrzQIY23gr82cy6Ah/m6Esq+HkDuAfIGWyZ2W3AYuAYMzsmBpCpUawzqnEcAJB0gaSZkmau+GxZriY4t0Xbfffdee+995gzZw4jRozgJz/5CZ9//jmlpaUsWLCAU045Jef2Z511FnPmzOGhhx5ixIgR9OvXjyeffJJevXoxYMAAvvrqq1rqiWuoPICqRyQVEabT+pvZ5zH5z8DeQCdgCXBTqniGXRhwFPBPMyuPo1mP5ahyKmEkad9E2neBQ4EZkkrj67YxrzK2L2l04vk9wE/jSM3pwN+A/YF34igZwMOJ8kcBDwKY2XPAzpJ2SNv/kYltHszRl1Twsz9hSvMBSZmOUb4KPQ4AmNldZtbFzLrssOPOG1G9cw1b06ZN2Xnn8Bk59NBD2XvvvXnrrbd46aWXmDVrFm3atOGoo47irbfeonv37ln3s3jxYmbMmEHPnj0ZNmwYo0ePpmnTpjz77LO11BPXUPnPGNQTkpoQLsqjzGxsKt3MPkqUuRtITUctAlondlFMGFH5ZgHVTgbuB56U9G0zW0wIzO43s99lKL/azCrT0lYlnj8KDAWeI6ypWiYpV3uyBYH5pGVlZi9J2oUw2rWW9b9INMtzN4UeB+dcAZYuXcpOO+1E48aNeeedd5g/fz5t27alS5cu/PKXvwTCnXonnXQSkyZNyrqfwYMHc801YcC5vLwcSTRq1IiysrLa6IZrwHwEqh6IIyX3Aq+b2Yi0vORKzFOAV+PzCUBvSU0l7QXsA0wnjCqdHO+sKwJ+kKtuM3sUGA48JakF8CzQS1LLWP9OkvbMpx9mthp4mjBq9teY/AbQNk5NQhiZSplMnBaT1B34JDHylvIfwmJ5+HoKLSdJ+wONgWXA/4AD43HagTCSlPIFsF3idUUMZGEjjoNzW7pBgwZRXFxMWVkZxcXFlJSUADBhwgSGDBkCwOTJk+nQoQMdO3akV69e3Hnnney0004F1TNnzhwAOnfuDMB5553HwQcfzOzZszn++OM3XYfcFklmBX15d3VA0lHAFGAekJq4v9zMJkp6kDB9Z8BC4BdmtiRudwXwM8IoS38zezKmlwB9CMHDUmCSmd2dVmcJsNLMbky8/g7wPcIC9d8RAvAK4Fdm9rKklWZWlNjHQqCLmX2SSOtGGIn6ZmqURtLJhCDtE0KQt1tcZ7QTIdDaCygDLjCzuZLOjfvtG4PDvxFGUx8Frky2IVFvZTx+EEaPLjezJ2LeDbFP84E1wAQzGynp18CvgCVmdoyk64EewOzYvtPzOQ7ZtGvf0W549Kmqijm32fEf0nR1SdIsM8u1frd22uEB1JZHUpGZrVT4jaPJhMBkdlXbbaK6BwI7mNngRFqqPQL+SLjr7+baaE9d8gDK1VceQLm6tLkEUL4Gast0l8LvRDUjrOOpreBpHGHB+3fSsn4u6Rxga2AO4a68Bq9FsyZ+IXLOuXrKA6gtkJn9pI7qzXjfcRxtavAjTs455xoOX0TunHPOOVcgD6Ccc8455wrkAZRzzjnnXIE8gHLOOeecK5AHUM4555xzBfIAyjnnnHOuQB5AOeecc84VyAMo55xzzrkCeQDlnHPOOVcg/yVy5+rI8tUVjH1zSV03wznA/76dc4XyESjnnHMb5YorrqB169YUFRVVWfa9996jqKiIG2+8cb30yspKOnfuzEknnZRxu8suu4wOHTpw9tlnr0t78MEHufXWWzeu8c5VU60EUJLuk/SxpFfT0odLekPSXEnjJLWI6W0klUsqjY87C6zvXEkm6buJtFNiWq/4epKkNyW9Iuk/kvZLpL8nSYltx0tamaWuytjGVyTNlvStmN5d0uNpZUdmqD/VxzExvUTSB4n067KU75VWf+rx20T5Lmn1d4/H4LxEWueYNjC+lqQrJc2X9Jak5yUdFPMOT6urVNJqSb+Mxzc97ytJJyTqGhDL75DWphWS5sT3wo2JvHMlLU3b54Hx/bHeeyn9+GbI26DuHOeyVNKERPqoeOxfje/lJjH9VEmvSZoiaeeYtrekR7LV4VxDdPLJJzN9+vS8yg4YMIATTjhhg/Rbb72VAw44IOM2K1as4MUXX2Tu3LlUVlYyb948ysvLGTlyJBdddNFGtd256qqtEaiRwPEZ0p8B2ptZB+At4HeJvLfNrFN8XFiNOucBfRKvewOvpJU5w8w6AvcDwxPpy4EjAWJQl2tsuzy2sWNs/7UFtPGMRB+TF/6bE+m/zVJ+TFr9qcd1VdQ5Dzg98Tr9uPwK+BbQ0cz2jf2ZIKmZmU1L1gUMAhYCD5jZuLS8PwFTgKcT++4DzADS/6jwFDPrDHQGTpJ0ZCJvdFr//ltF/7LJVndS8lj2SKSPAvYHDga2Ac6P6ZcA3YAHgNQfaB4GDK5mG52rl7p168buu1c9BTh+/Hjatm3LQQcdtF76okWLeOKJJzj//PMzbteoUSPWrFmDmVFeXk6TJk0YPnw4/fr1o0mTJpukD84VqlYCKDObDHyaIf1fZrY2vnwZKK5qX5LOiyMjkyTdLemOLEWnAIdJaiKpCGgHlGYpOznmpzxCCCwAfgSMrapd0fbAZ3mWrSvvAc0k7RZH2Y4HnkzkXwb82szKIJwj4EXgjOROJO0C3E0I6lal5e0LDAHOMrOvYtreQBFwJesHtuuYWTnhHO2xsZ1Ma0+VdediZhMtAqbz9fv0K6Ap0ByokPRtYImZzd80LXeu4Vi1ahXXX389Q4cO3SCvf//+3HDDDTRqlPmStN1223HqqafSuXNn9tprL3bYYQdmzJhBz549a7rZzmW1Oa2B+hnrX8j3itM6L8QLE5JaEb7ddwOOI4wKZGPAv4HvAz2BCTnKnkwYmUl5Fvg/SY0JgdToHNtuE6d83gDuAa7JUTbdqMSUUXIEbEAi/ftZyu+cVn/qkRxdymYM8GPCSNNs4EsASdsD25rZ22nlZwIHpaXdC/zJzGYlE+P01t+AgWb2XiKrD/AwIbDdT1LL9EZJ2hHYhxDQppye1r9t8uhfuirrjppJminpZUk/zNC+JsBZwFMx6SrCCNuxcf9XUsX5l3RBrGPmis+WVaMrztVPQ4cOZcCAARusk3r88cdp2bIlhx56aM7tBw0aRGlpKTfddBODBw/m6quv5p577uG0005j2LBhNdl05zLaLO7Ck3QFsJYwVQKwBPimmS2TdCgwPq7DOQx4wcw+jdv9A9g3x64fAfoBOxCmWy5Pyx8lqZwwDfXrRHolMJUw1bWNmS3U10ui0pXHKSskHQE8IKk9IYDLJJl+hpnNzFDmZjO7MUN6pvLr6i/A3wlB4f6EC/+3qigvEu2WdCFhtG14hrLXAK+ZWfo6oN7AKWb2laSxhADujzHv25LmAvsB15nZh4ntRptZ3/Uak/1cZJOr7qRvmtliSW2B5yTNSwsm/wRMNrMpAGb2DGEaGknnABMJAdpAwkjkxamRvBQzuwu4C6Bd+47Z3iPONTjTpk1jzJgxDBo0iOXLl9OoUSOaNWvGBx98wIQJE5g4cSKrV6/m888/58wzz+Shhx7KuJ85c+YAsO+++3LxxRczefJkevfuzfz589lnn31qs0tuC1fnAVS88JwEfDdOkWBmXxJHRcxslqS3CYFSQVdOM5seg5lyM3srw4U3WwADIfgaB5QUUN9LcWprV2AZsGNakZ2AT/LdX00xsw8lVRBG8S4mBlBm9rmkVZLamtk7iU0OAV4AkLQ/YaSlW2p6LkVSd+DUWD6Z3oEwsvRMPAdbA+/wdRAzxcxOilN/UyWNM7Ns060FyaPudcxscfz3HUmTCGuy3o77GUo4r7/IUEdz4BzCaOe/CCOePyFMe969KfrhXH03ZcqUdc9LSkooKiqib9/w3ejaa8PS0UmTJnHjjTdmDZ4ABg8ezF133UVFRQWVlZVAWCNVVlaWdRvnakKdTuFJOp6w5qZH8pu6pF3j9BlxNGAfwkVvOnC0pB0lbUW4WFfld2w48pSPKYQF1A/nu0EMLhoTgqf5QCtJB8S8PYGOZF+HVduGAJeZWWVa+nDgttRUmaRjgaOAv0namjA9N8DMFiU3itNvfwXONrMv0vbZBygxszbx0QrYIx6TdczsLcIxv2zTdDH/uuN7qml8vgvhJoL/xtfnE4KjPulBYzQIuNXMKgiLzI2wPqr5JuyHc5utQYMGUVxcTFlZGcXFxZSUlAAwYcIEhgwZssnqGT9+PF27dqVVq1a0aNGCI444goMPPhhJdOzYcZPV41w+FAd9arYS6WGgO7AL8BEw1MzulbSAsAg3tRjkZTO7UNKpwNWEab3KWP6xuK8LgIHAYuB14FMzuyKtvnOBLhmmfkYCj5vZmDjCMDB9BCpH+koz2+BHTiRV8vX6KQGXm9kTMe9I4CagGVAR81JTPpMId/eVx20/MbNjJZUAK9On8HK0K1k/wFNm9ttY/oBYL8BLhFGXgWZ2Uto+1tUZF5YPIaz1qQQ+BPqa2TxJfQh3nL2Wdhjuj328khA4Jl0LXAecYGZvJOocQXgvTEu2KQZuCwhB29GEgO6DxP4uIpz7+XH7lAHADwjr2VLH9H3gG9nqNrPrE2nfAv5CCHwaAbeY2b0xby3wPyAVGI41s6tjXivgrkT7f0wYtVwO/NDMlpJFu/Yd7YZHn8qW7Vyt8h/SdPWFpFlm1qXqkjXcjtoIoDYlSUVmtjKOQI0D7jOzcXXdLucK5QGU25x4AOXqi80lgKrzNVDVUBKnlZoR1puMr+P2OFctLZo18YuWc87VU/UugDKzgXXdBuecc85t2Tan34FyzjnnnKsXPIByzjnnnCuQB1DOOeeccwXyAMo555xzrkAeQDnnnHPOFcgDKOecc865AnkA5ZxzzjlXIA+gnHPOOecKVO9+SNO5hmL56grGvrmkrpvhnP8ivnPV4CNQzjnnnHMF8gDKOedctV1xxRW0bt2aoqKiKsu+9957FBUVceONNwKwevVqDjvsMDp27MhBBx3E0KFDM2532WWX0aFDB84+++x1aQ8++CC33nrrpumEc9XgAVQkaaGkeZJKJc1MpE+StFF/9VnSPyW9lJZWIumDWN9/JfVJ5EnSlZLmS3pL0vOSDsqy70mS3oz7KZXUK6ZXxtevSnpMUouY3kbSq/F5c0mjYr9flTRVUlHM+4akRyS9Hds3UdK+ye3T+jIwPh8p6d1Ee17M0Oas9cb8UySZpP0TafnU+4GkpvH1LpIWJrYtT7SpVNLZpJF0r6RXJM2VNCZxLJpKGi1pgaRpktrE9CNj2RmS2sW0FpKelqRM58u5hubkk09m+vTpeZUdMGAAJ5xwwrrXTZs25bnnnuOVV16htLSUp556ipdffnm9bVasWMGLL77I3LlzqaysZN68eZSXlzNy5EguuuiiTdoX5wrha6DWd4yZfbIpdxgDl0OAlZL2MrN3E9k3m9mNkvYBZkkaY2YVwK+AbwEdzaxM0veACZIOMrPVGao5w8xmpqWVm1mn2Ib74z5/n1bmYuAjMzs4ltsPqIgX/3HA/WbWO+Z1AnYD3s+j25ea2Zgc+RnrTeT3AaYCvYGSPOpLqQR+Bvw5Q97bqeORwwAz+zy2aQTQF7gOOA/4zMzaSeoNXA+cDlwCnAq0AX4ZXw8G/mBmVkC7nau3unXrlle58ePH07ZtW7bddtt1aZLWjVxVVFRQUVFB+nePRo0asWbNGsyM8vJymjRpwvDhw+nXrx9NmjTZdB1xrkA+ApUnSY0k3S9pWHx9XhwdmiTpbkl3ZNn0VOAx4BFCQLABM5sPlAE7xqTLgF+bWVnM/xfwInBGNZv/ErBHhvTdgQ8S7XjTzL4EjgEqzOzORF6pmU2pZv351ksc9TmSELRkPF453AIMkFStLwaJ4EnANkAqCOoJ3B+fjwG+G8tUxHLNCYHn3sAeZvZCdep3rqFatWoV119/fcYpusrKSjp16kTLli057rjjOPzww9fL32677Tj11FPp3Lkze+21FzvssAMzZsygZ8+etdV85zLyAOprBvxL0ixJF6TlbQWMAt4ysysltSKMNHQDjgP2J7s+wMPx0SdTAUmHAPPN7GNJ2wPbmtnbacVmAhmn8YBRiampndP23Rj4LjAhw3b3AZdJeknSsDgSBtAemJWjT3snp8OAC9PyhyfyRxVQL8APgafM7C3g03hs8q33PcLI1VlVtVnStzN1TNJfgQ8J5/T2mLwHceTNzNYCK4CdgWuBu4D+wB2EEb7Bmfab2P8FkmZKmrnis2W5ijrXYAwdOpQBAwZkXCfVuHFjSktLWbRoEdOnT+fVV1/doMygQYMoLS3lpptuYvDgwVx99dXcc889nHbaaQwbNqw2uuDcBnwK72tHmtliSS2BZyS9YWaTY95fgL+bWWoK7DDgBTP7FEDSP4B903coaTegHTDVzEzSWkntzSz1P8QAST8H2gLHV9E+8fWISLpMU3jbxCCjDSEYeiZ9IzMrldQW+B5wLDBD0hFVtAPSpsMklaTl55zCy1avmb1OCDJviUUfia9n51kvwB8IweITudqco20/jUHn7YRpur8Sjn2GolZKCKKR9H/A4vBUowmjU5eY2UdpG91FCLpo176jT/O5LcK0adMYM2YMgwYNYvny5TRq1IhmzZrRt2/fdWVatGhB9+7deeqpp2jfvn3G/cyZMweAfffdl4svvpjJkyfTu3dv5s+fzz777JNxG+dqio9ARWa2OP77MWH9z2GJ7BeBYyQ1i6/zXSB8OmFa7t24oLkN609L3Wxm+8VyD0hqFqeRVsUAI+kQ4L/592jdGqg9ga0Ja6A2YGYrzWysmV0EPAScCLwGHFpAXQXLVG8cPfsOcE88XpcCpxeyINvMFgClwGkb0bZKYDRh+hVgEdAaIE4P7gB8miof23clcA0wND4eAvpVtw3ONSRTpkxh4cKFLFy4kP79+3P55ZfTt29fli5dyvLlywEoLy/n3//+N/vvn31APzX6VFFRQWVlJRDWSJWVldVKP5xL8gAKkLStpO1SzwkjI8lx5HuBicA/4gV0OnC0pB3j61PT9xn1AY43szZm1oYQlGywrsfMxhKm6M6JScOB2yRtE9t0LHAU8LdC+2ZmKwgX8oGS1ltxGe8i2zE+3xo4EPgf8BzQNI6Opcp2lXR0ofVnkqPeXsADZrZnPGatgXcJfS/E74GBBbZJiTvpBJwMvBGzJ/D1uekFPJe2SPwc4Akz+4ywHuqr+GheYLudq3cGDRpEcXExZWVlFBcXU1JSAsCECRMYMmRIzm2XLFnCMcccQ4cOHejatSvHHXccJ510Usay48ePp2vXrrRq1YoWLVpwxBFHcPDBByOJjh07bupuOVcl+c1CEEd7xsWXWwF/S03XSZoEDDSzmZKuIkzVnQGcT7hILwZeBz41sysS+2wD/AcoTl5sJc0m3LF1ArDSzG6M6YcSAqQDCFN1QwhreSoJa3L6mtm8DG1f17609JVmlvxpgMeAvwNTgMfNrL3CrfwDCSNqjQjTXpfF6cZWhKm0Q4HVwELCWp+K1PaJfZek+iJpJHA0YZ1QymFmtiZRPmO9wPPAdWb2VKJsv3hMrs+j3sdTU4eSxgKHmFmbeC5eB95MtOk+M7stsa9G8dhsH9v1CvBLM/s8jjw+CHQmjDz1NrN34nbNY/u/Z2YVcW3Vn4A1QJ+4liujdu072g2PPpUt27la479E7uoTSbPMbKN+XmiTtMMDqOqRVGRmK+MI1DjCBXlcVds5l9KlSxebOTN96ZpzzrlcNpcAyqfwqq8kLtJ+lTDNNL6O2+Occ865WuJ34VWTmRW0xsY555xzDYePQDnnnHPOFcgDKOecc865AnkA5ZxzzjlXIL8Lz7k6IukL1v9phYZoF2CT/oHuzZD3sWHwPtYfe5rZrnXdCF9E7lzdeXNzuBW3Jkma6X2s/7yPDcOW0Mfa5FN4zjnnnHMF8gDKOeecc65AHkA5V3fuqusG1ALvY8PgfWwYtoQ+1hpfRO6cc845VyAfgXLOOeecK5AHUM4555xzBfIAyrkaJul4SW9KWiDptxnym0oaHfOnSWpT+63cOHn08VxJSyWVxsf5ddHO6pJ0n6SPJb2aJV+Sbov9nyvpkNpu48bKo4/dJa1InMMhtd3GjSWptaTnJb0u6TVJF2coU6/PZZ59rPfncnPgvwPlXA2S1Bj4I3AcsAiYIWmCmf03Uew84DMzayepN3A9cHrtt7Z68uwjwGgz61vrDdw0RgJ3AA9kyT8B2Cc+Dgf+HP+tT0aSu48AU8zspNppTo1YC1xiZrMlbQfMkvRM2nu1vp/LfPoI9f9c1jkfgXKuZh0GLDCzd8xsDfAI0DOtTE/g/vh8DPBdSarFNm6sfPpYr5nZZODTHEV6Ag9Y8DLQQtLutdO6TSOPPtZ7ZrbEzGbH518ArwN7pBWr1+cyzz66TcADKOdq1h7A+4nXi9jwP7N1ZcxsLbAC2LlWWrdp5NNHgFPjlMgYSa1rp2m1Jt9jUN8dIekVSU9KOqiuG7Mx4lR5Z2BaWlaDOZc5+ggN6FzWFQ+gnKtZmUaS0n87JJ8ym7N82v8Y0MbMOgD/5usRt4aivp/DfMwm/A2yjsDtwPg6bk+1SSoCHgX6m9nn6dkZNql357KKPjaYc1mXPIByrmYtApKjLcXA4mxlJG0F7ED9mkqpso9mtszMvowv7wYOraW21ZZ8znO9Zmafm9nK+Hwi0ETSLnXcrIJJakIILEaZ2dgMRer9uayqjw3lXNY1D6Ccq1kzgH0k7SVpa6A3MCGtzATgnPi8F/Cc1a9fuK2yj2lrSHoQ1mU0JBOAs+MdXN2AFWa2pK4btSlJ+kZa0qEPAAALvUlEQVRqbZ6kwwjXj2V126rCxPbfC7xuZiOyFKvX5zKfPjaEc7k58LvwnKtBZrZWUl/gaaAxcJ+ZvSbpamCmmU0g/Gf3oKQFhJGn3nXX4sLl2cd+knoQ7hD6FDi3zhpcDZIeBroDu0haBAwFmgCY2Z3AROBEYAFQBvy0blpafXn0sRfwS0lrgXKgdz0L9AGOBM4C5kkqjWmXA9+EBnMu8+ljQziXdc7/lItzzjnnXIF8Cs8555xzrkAeQDnnnHPOFcgDKOecc865AnkA5ZxzzjlXIA+gnHPOOecK5AGUc84551yBPIByzjnnnCuQB1DOOeeccwXyAMo555xzrkAeQDnnnHPOFcgDKOecc865AnkA5ZxzzjlXIA+gnHPOOecK5AGUc84551yBPIByzjnnnCuQB1DOOeeccwXyAMo555xzrkAeQDnnnHPOFcgDKOecc865AnkA5ZxzzjlXIA+gnHPOOecK5AGUc67WSHpNUvdarlOS/irpM0nTC9zWJLUroPwZkv6VZ9lzJU0tpD01QdKRkuZLWinphzVUx0JJx9bEvrdUm8v7Z0vmAZRz9Vi8MJXHi99HMVAoqut2AUgaKWlYMs3MDjKzSbXclKOA44BiMzusJisys1Fm9r2arKMGXA3cYWZFZjZ+Y3eW6bxvSoUGtQXuu72kpyV9Isky5O8kaZykVZL+J+knNdGO2rSpgltJO8Rjt1zSKEmNE3l3SzplY+vY3HgA5Vz9d7KZFQGHAF2BK9MLxFGYWvu8J//z3AzsCSw0s1U1WYmkrWpy/zVoT+C16mxYj/ucTQXwd+C8LPl/BNYAuwFnAH+WdNDGVtpAjuMvgDmEY9MGOAVA0hHA7mY2ru6aVjM8gHKugTCzD4AngfYAkiZJ+r2k/wBlQFtJrSRNkPSppAWSfp7aXlKJpDGSRkv6QtJsSR0T+QfEfS6PU3E9EnkjJf1Z0kRJqwgXoDOAQXF07LFYbt23XUlNJd0iaXF83CKpaczrLmmRpEskfSxpiaSfZut7tn5JOg+4BzgituOqDNu2k/SCpBVx5GF0WpFj4xTXZ5L+KElxu3Ml/UfSzZI+BUrSp1XiaMmFmbbP0I7hkqbGb/J7S3pO0rLYplGSWiTKHiJpTjxP/4jnbFgi/yRJpfFcvSipQ5Y63wbaAo/F49M0z/fIQ5I+B85N298FZDjvUSdJc+NxHi2pWTXaOzk+fSXu//SY/vPY1k9j21ulnYN+kt6Jx3K4snyZMLM3zexeMgSUkrYFTgUGm9lKM5sKTADOytLWqj5PCyVdJmkusErSVlV8xnaOfftcYSp670Rem9jPrRJpkySdn3j9c0mvx7b8N76HHgS+ydfnf5CkZvH8LovtmCFpt0x9TLMX8LyZfQlMIfx/0xi4Gbg4j+3rHzPzhz/8UU8fwELg2Pi8NeE//mvi60nAe8BBwFZAE+AF4E9AM6ATsBT4bixfQvgG3iuWHQi8G583ARYAlwNbA98BvgD2i9uOBFYARxK+mDWLacNytPdq4GWgJbAr8GKi7d2BtbFME+BEQhC4Y5bjkKtf5wJTcxzDh4ErEu0+KpFnwONAC8KFZilwfGK/a4Ffx+O7TXpdeWw/NdZ7N/A00DzmtSNMOzaNx2YycEvM2xr4H+Gi1AT4EWFUZFjMPwT4GDgcaAycE49706reQ3kcyxLCe+SHsd3bZNhftvM+HWgF7AS8DlxYzfYa0C7x+jvAJ3E/TYHbgclp5Z+P9X4TeAs4v4rPVTvA0tI6A+VpaQOBx7LsI3WsNvg8JY5JKeFzuw1Vf8YeIYyObUv4kvQB8b1GGPExYKtE/ZNS/QR+HMt3BRT7t2eW8/8L4DGgeTwfhwLb5/F/0a+A4bEv/wF+AAwAhtb1/5M19fARKOfqv/GSlhMuxi8Af0jkjTSz18xsLfANwnqgy8xstZmVEkZnkt+gZ5nZGDOrAEYQLqLd4qMIuM7M1pjZc4TAoE9i23+a2X/M7CszW51Hu88Arjazj81sKXBVWlsqYn6FmU0EVgL7pe9EUus8+pVLBWEaq1XcPn1h7nVmttzM3iNciDsl8hab2e1mttbMyrPsP9f2TQgB3E6EqdgyADNbYGbPmNmX8diMAI6O23QjBGy3xWMzlhCcpPwc+IuZTTOzSjO7H/gybpdTnsfyJTMbH89ztj5ncpuZLTazTwkX6NRxqHZ7ozOA+8xstoXRj98RRhzbJMpcb2afxnNwC+u/b/NVRPiSkLQC2C7HNtk+Tym3mdn78Thm/YzFkZxTgSFmtsrMXgXuL6Dt5wM3mNkMCxaY2f+ylK0AdiYEqZVmNsvMPs+jjnuBHYBphBGoVwjvm1sURqcnqwbXxtUFD6Ccq/9+aGYtzGxPM7so7aL2fuJ5K+BTM/sikfY/YI9M5c3sK2BR3K4V8H5Mq3LbPLWK+0jur1Xi9bIY+KWUES4wmfZTVb9yGUT4Vj49Tpv8LC3/wxxtyKfPubZvB/QErjKzNalESS0lPSLpgzhV9hCwS8xuBXxgZslFzsl27AlcEqdflsfgujXrH9tsCnqPFCjbcdiY9qbavO59ZGYrgWU52pz+PsvXSmD7tLTtCaNE2WT7PGVqV67P2K6EoDm9H/lqDbydZ9kHCaOhjyhMrd8gqUlVG8WA+wIz62BmvyVM3V1OCHAbE74AHC7p+ALavVnzAMq5hi15kV0M7CQp+Y35m4Sh/ZTWqSdxnUhx3G4x0Dpt7Uj6tul3LW1wF1OaxYSLZ3J/i6vYJtt+qupXVmb2oZn93MxaEaYv/qT87/Kqqo9VeR34KfCkpOTo2rVx3x3MbHvgTEKQB7AE2ENaby1V68Tz94Hfx6A69WhuZg/n0Z58jmVVfS70mGxMeyHtfRTXKu1Mlvc11X+fvQVsJWmfRFpHci/Az/Z5Skn/fGb7jC0lTBen9yMldYNE80TaNxLP3yexZirNeucrjmpeZWYHAt8CTgLOzrJtRjFIkpk9BRwMzIwB/0wg4/q2+sgDKOe2EGb2PmGd0bVxoWgHwmLvUYlih0r6UVyM2p8wlfIyYVh+FWFxcBOF33I6mbAuI5uPCAuUs3kYuFLSrpJ2AYYQRlpqol9ZSfqxpOL48jPCBaWy0HZUVwwULgf+LSl1kduOMOKxXNIewKWJTV6K7esbFx73BJI/z3A3cKGkwxVsK+kHaUFRtrZs1LGMqjrv6Qptb/r+/wb8VFInhZsQ/gBMM7OFiTKXStoxTlFeDKTfKACsu1u1GWENEvEYNAWwcBfnWODq2MYjCaOHD+boW7bPUyZZP2NmVhnrLpHUXNKBhLVixLYtJQRaZ0pqHEdRkwHTPcBASYfGPraTlAo61zueko6RdHCcNvycMKVXGfNKJE3K0V/i8buOsP4Jwrqv7pK2JqyRfCfX9vWJB1DObVn6EBacLgbGERZ4PpPI/ydwOiGQOAv4UfxGugboAZxAWLD7J+BsM3sjR133AgfGaZlMvy80jPCNdC4wD5gd02qiX7l0BaZJWkm4q+piM3u3mu2olrju52rgubh25yrCougVwBOEi2eq7BrCwvHzgOWE0anHCRdnzGwmYV3RHYTzuIC0u+WqsDHHEqo+7+upRntLgPvj/k8zs2eBwcCjhNG5vYHeadv8E5hFWLT9RGxjJnsC5Xw9qlQOvJnIv4iwSPpjwheAX5pZrhGojJ+nTAXz+Iz1JUx7fkhYqP/XtF38nBBoLyPcOPJiYt//AH5PCDa/AMYT1t1BGO28Mh7PgYSRqzGE4Ol1wrrK1Beb1oQF4rlcDoyKwTjAXwjTz0sJU5gN5ucMtP40unNuSyWphLBw9My6bosrjKRpwJ1mln5R3eIp/CDmPma2oJbrLaGBfZ4klRLuyFxW123ZHDSEH+9yzrktiqSjCSMjnxAW6XYAnqrTRrkGz8w6VV1qy+EBlHPO1T/7EX4TqIhwd1UvM1tSt01ybsviU3jOOeeccwXyReTOOeeccwXyAMo555xzrkAeQDnnnHPOFcgDKOecc865AnkA5ZxzzjlXoP8HhyWjd+b86zsAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "df_by_name_top_shrinkage = df.groupby(\"article_name\")[\"daily_shrinkage_in_euros\"].sum().reset_index()\n",
    "df_by_name_top_shrinkage[\"% out of total shrinkage\"]= (df_by_name_top_shrinkage[\"daily_shrinkage_in_euros\"]/df_by_name_top_shrinkage[\"daily_shrinkage_in_euros\"].sum())* 100\n",
    "df_by_name_top_shrinkage = df_by_name_top_shrinkage.nsmallest(10, [\"daily_shrinkage_in_euros\"])\n",
    "df_by_name_top_shrinkage.set_index(\"article_name\", inplace=True, drop=\"true\")\n",
    "\n",
    "ax = df_by_name_top_shrinkage[\"% out of total shrinkage\"][:10].plot.barh(color=\"lightblue\")\n",
    "plt.xlabel(\"Proportion of shrinkage for the top 10 products, %\", fontsize = 12, labelpad=20);\n",
    "plt.ylabel(\"Article name\", fontsize = 12);\n",
    "for patch in ax.patches:\n",
    "    ax.text(\n",
    "        patch.get_width() + 0.1, \n",
    "        patch.get_y() + 0.38,\n",
    "        \" {} %\".format(patch.get_width().round(2)), \n",
    "        fontsize=10,\n",
    "        color='black'\n",
    "    )\n",
    "\n",
    "ax.invert_yaxis()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Figure 7. Top 10 products with the highest shrinkage in the department store for January 2020*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Based on the shrinkage of products across different categories in the department store, the top 10 products with highest shrinkage are shown on Figure 7:\n",
    "\n",
    "1. 5 kg Frischer Kloßteig\n",
    "2. 10er Eier weiss/braun L-M\n",
    "3. 500g ARO RAEUCHERLACHS\n",
    "\n",
    "...\n",
    "\n",
    "These 10 products make up around 18% of total shrinkage. \n",
    "Let's explore only the first three products with the highest shrinkage and find out what is the proportion of these products in sales."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "                   article_name  daily_sales_in_euros\n",
      "7257  5kg FRISCHER KLOSSTEIG BY                 192.0\n",
      "                       article_name  daily_sales_in_euros\n",
      "815  10er Eier weiss/braun L-M Bode                2026.0\n",
      "                    article_name  daily_sales_in_euros\n",
      "6248  500g ARO RAEUCHERLACHS (2)                1869.0\n"
     ]
    }
   ],
   "source": [
    "df_new=df.groupby(\"article_name\")[\"daily_sales_in_euros\"].sum().reset_index()\n",
    "top_3 = [\"5kg FRISCHER KLOSSTEIG BY\",\"10er Eier weiss/braun L-M Bode\", \"500g ARO RAEUCHERLACHS (2)\"]\n",
    "for product in top_3:\n",
    "    print(df_new.loc[df_new[\"article_name\"] == product])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.5811589207849548\n"
     ]
    }
   ],
   "source": [
    "print(sum([192, 2026, 1869])/df_new[\"daily_sales_in_euros\"].sum() *100)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Top 10 best-selling products in the department store for January 2020 make almost 11% of total \n",
    "\n",
    "As a conclusion, the top 3 products with the highest shrinkage take around 7% out of total shrinkage and make 0,58% of total sales. As a suggestion to \"Metro\", to check the reasons for such a high shrinkage of these products and these products could be removed from the store."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Exploring the main category of products"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['ACM FLEISCH',\n",
       " 'ACM BACKWAREN',\n",
       " 'ACM FRISCHFISCH',\n",
       " 'ACM MOLKEREIPRODUKTE',\n",
       " 'ACM KAESE',\n",
       " 'ACM FEINKOST',\n",
       " 'ACM GOURVENIENCE',\n",
       " 'ACM WURST',\n",
       " 'ACM OBST',\n",
       " 'ACM GEMUESE']"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df[\"purchasing_area_description\"].unique().tolist()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "AMC BLUMEN has no any record."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>purchasing_area_description</th>\n",
       "      <th>daily_sales_in_euros</th>\n",
       "      <th>daily_shrinkage_in_euros</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>ACM BACKWAREN</td>\n",
       "      <td>14102.000</td>\n",
       "      <td>-199.9313</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>ACM FEINKOST</td>\n",
       "      <td>25170.000</td>\n",
       "      <td>-3371.1331</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>ACM FLEISCH</td>\n",
       "      <td>123057.865</td>\n",
       "      <td>-2184.6305</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>ACM FRISCHFISCH</td>\n",
       "      <td>28823.007</td>\n",
       "      <td>-1256.6433</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>ACM GEMUESE</td>\n",
       "      <td>102817.000</td>\n",
       "      <td>-2363.5960</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>5</td>\n",
       "      <td>ACM GOURVENIENCE</td>\n",
       "      <td>2502.810</td>\n",
       "      <td>0.0000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>6</td>\n",
       "      <td>ACM KAESE</td>\n",
       "      <td>125558.000</td>\n",
       "      <td>-3716.3172</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>7</td>\n",
       "      <td>ACM MOLKEREIPRODUKTE</td>\n",
       "      <td>164432.299</td>\n",
       "      <td>-4353.7868</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>8</td>\n",
       "      <td>ACM OBST</td>\n",
       "      <td>54486.000</td>\n",
       "      <td>-230.1300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>9</td>\n",
       "      <td>ACM WURST</td>\n",
       "      <td>62301.000</td>\n",
       "      <td>-5156.0581</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  purchasing_area_description  daily_sales_in_euros  daily_shrinkage_in_euros\n",
       "0               ACM BACKWAREN             14102.000                 -199.9313\n",
       "1                ACM FEINKOST             25170.000                -3371.1331\n",
       "2                 ACM FLEISCH            123057.865                -2184.6305\n",
       "3             ACM FRISCHFISCH             28823.007                -1256.6433\n",
       "4                 ACM GEMUESE            102817.000                -2363.5960\n",
       "5            ACM GOURVENIENCE              2502.810                    0.0000\n",
       "6                   ACM KAESE            125558.000                -3716.3172\n",
       "7        ACM MOLKEREIPRODUKTE            164432.299                -4353.7868\n",
       "8                    ACM OBST             54486.000                 -230.1300\n",
       "9                   ACM WURST             62301.000                -5156.0581"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_areas = df.groupby(\"purchasing_area_description\")[\"daily_sales_in_euros\",\"daily_shrinkage_in_euros\"].sum().drop(index=\"ACM BLUMEN\").reset_index()\n",
    "df_areas"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Let's explore purchasing areas in depth."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Shrinkage in EUR by purchasing areas')"
      ]
     },
     "execution_count": 42,
     "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": [
    "sns.barplot(data = df_areas, x = \"purchasing_area_description\", y = \"daily_shrinkage_in_euros\")\n",
    "plt.xticks(rotation=-90)\n",
    "plt.xlabel(\"Purchasing areas\",  fontsize = 12, labelpad=20);\n",
    "plt.ylabel(\"Total shrinkage, EUR\", fontsize = 12);\n",
    "plt.title('Shrinkage in EUR by purchasing areas', pad=20,color=\"black\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Figure 8. Total shrinkage in euros by purchasing area*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Areas with the highest shrinkage (following the Figure 8): ACM WURST, ACM MOLKEREIPRODUKTE and ACM KAESE, areas with the lowest shrinkage is ACM OBST and ACM BACKWAREN. \n",
    "ACM  GOURVENIENCE does not have shrinkage. It is the own brand of \"Metro\", so it could be that the shrinkage counts in a different way."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Sales in EUR by purchasing areas')"
      ]
     },
     "execution_count": 43,
     "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": [
    "sns.barplot(data = df_areas, x = \"purchasing_area_description\", y = \"daily_sales_in_euros\")\n",
    "plt.xticks(rotation=-90)\n",
    "plt.xlabel(\"Purchasing areas\",  fontsize = 12, labelpad=20);\n",
    "plt.ylabel(\"Total sales, EUR\",  fontsize = 12);\n",
    "plt.title('Sales in EUR by purchasing areas', pad=20,color=\"black\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Figure 9. Total sales in euros by purchasing area*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The areas with the highest sales(following Figure 9): ACM MOLKEREIPRODUKTE, ACM KAESE and ACM FLEISCH and with the lowest sales ACM BACKWAREN and ACM GOURVENIENCE which is similar to shrinkage.\n",
    "\n",
    "Let's explore sales by category by plotting boxplot."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 44,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0, 400)"
      ]
     },
     "execution_count": 44,
     "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": [
    "sns.boxplot(x = df[\"purchasing_area_description\"], y = df[\"daily_sales_in_euros\"])\n",
    "plt.xticks(rotation=-90)\n",
    "plt.xlabel(\"Purchasing areas\",  fontsize = 12, labelpad=20);\n",
    "plt.ylim(0,400)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Figure 10. Statistical visualisation of the sales by the main categories of products*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Using statistical visualisation of the sales by the main categories of products (Figure 10) we found that we have a lot of outliers. After the research of the topic, it was clarified that it is normal for the real data sets. Here we can see that median is the best average measure for our data set. We can infer from Figure 10 that the highest average of total sales takes the ACM FLEISCH category(63 EUR). We can conclude, that despite the fact that ACM MOLKEREIPRODUKTE has the highest total sales,  ACM Fleish category has the biggest normalised max values."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Finding correlation"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The last step is to make a conclusion about correlation between the target variables."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(array([0.5, 1.5, 2.5, 3.5, 4.5, 5.5]), <a list of 6 Text yticklabel objects>)"
      ]
     },
     "execution_count": 45,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 600x600 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,10), dpi= 60)\n",
    "sns.heatmap(df.corr(), xticklabels=df.corr().columns, yticklabels=df.corr().columns, cmap='RdYlGn', center=0, annot=True)\n",
    "plt.xticks(fontsize=12)\n",
    "plt.yticks(fontsize=12)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Figure 11. The correllogram with correlation coefficients of all possible pairs of numeric variable*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Figure 11 doesn’t indicate that there’s any strong correlation, because the correlation coefficient of all possible pairs of numeric variables is less than 0.30. Correlation coefficient for sales in euros and shrinkage in euros is +0,025 which is very close to 0. We could prove the absence of correlation by plot (Figure 12) "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 46,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAa0AAAG6CAYAAAC7hRkUAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4xLjEsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy8QZhcZAAAgAElEQVR4nO3deZxbZ33v8c9Py6xextvEy4ztOIkxcUJCErIBjlugSSg4BQqFlK3QG4fSAqW3BcolpW65pWVpoQXqUHpZCgRCSmuWQAnBMVsWO4QkTpzEsWN7YsfjZWKPZ5f0u3/ojK0Zz6JZpKOj832/XnpJOudI+kkzo+88z3nOc8zdERERiYJE2AWIiIgUS6ElIiKRodASEZHIUGiJiEhkKLRERCQyUmEXME00BFJEqomFXUClUktLREQiQ6ElIiKRURXdg1+7Z29ZX+/6y5aW9fVERCRPLS0REYkMhZaIiESGQktERCJDoSUiIpGh0BIRkchQaImISGRUxZD3ifjV3g46uvs5u3kmLXPqSZgOPBcRiYpYhdav9nZw67Y2AO54tJ36dJKzmmewsnkG55wxk9n16ZArFBGRscQmtJ5o7+S2+9tYsaCR113cyu4jXTxx8AQ72zt5+OljADTPrOWcIMDOnN9IOqneUxGRShKL0Nr/bA9fvWcvzTPreONly6hLJ7mgpYkLWppwdw529vHEwU6eaD/BPbuP8vMnj5BKGMvnN3JO8wyeu2gW82fUhv02RERiz9yjP0H61+7ZO+qbONrVz8a7niSZMG686ixmjdMF2J/J8dSRrpMh1t7ZR8LgNRe18PylcwBN4yQiJaed7aOo6pZWV1+GL/5iN5mc8/YXnTluYAHUpBKsPGMmK8+YCcCz3f186/42bt3WxrGeAa5auaDUZYuIyCiqdqdNfybHV+7ew7PdA7zp8mU0z6qb1PM0NdTw1iuXc2FrE//zyEH+64H9ZLK5aa5WRESKUZWhlc0537hvL/uOdvO6S1pZPr9xSs+XSiR47cUtXLVyAfc9dZQbvrKN7v7MNFUrIiLFqrrQcne+8+v9PPpMJ6+4YDHnLZk9Lc9rZly9eiHXXbiYzY+18/qb7+ZQZ9+0PLeIiBSn6kLrJ48d4t6njnLVygVcsWLetD//ZWfO4+Y3XcLjBzt59ed+zq5DJ6b9NUREZGRVFVrb9hzljkcP8vzWJn7r3DNK9jovPfcMbrnhCrr7srzmc79g256jJXstERE5pWpC67FnOvn2r57mnOYZvPqiFqzE0zNd2NrEf/7RlcyuT3P95+/hBw8/U9LXExGRKgmtto5uvnbvHhbOquP6S5eSTJTnEIdl8xq57R1Xcu7iWbzjq9v44s93l+V1RUTiqipC60u/eIoZtSnecuVyatPJsr72vBm1fO0PL+elzz2DD3/nET7yvUfI5aJ/wLaISCWqitBy4K1XnsnMunAmvK2vSfKvb7yYN1+xjM//dDd/csuv6OnPhlKLiEg1q4oZMX7/smUsmBnu3IDJhPHX61azpKmev7t9B48908k//d6F0zbkXkREqqSldeYUDx6eLmbG+qvO4itvv5TjPQO86rM/51/velLdhSIi06QqQqvSvPicBfzwPWt4yaoz+OjtO7j+3+5m/7M9YZclIhJ5Cq0SmdNYw+feeBH/8Jrn8WDbMa75py1859f7wy5LRCTSFFolZGa87gWtfP9dL2bFghn8ydd/xXu/8QCdvQNhlyYiEkkKrTJYPr+RW2+8gne95Bz+64GnufZTP2XrU5pFQ0RkohRaZZJOJnjvy1Zy641XYAav2/hLPvE/jzGg05yIiBSt6s9cXIl6B7J898H93L/3WVrm1PO6S1qZP2P8Ifs6Y7JIbOjMxaNQSysEdekkv3txK2+4dClHTvTzL3fu5K7H2ukd0AHJIiJjqYqDi6Pq/CWzWTq3gf/61dP88JGDbH78EJevmMeVZ80LbXYPEZFKpu7BCvF0Rw93PXGI7U8fI5kwLlo6hxefM595Bd2G6h4UiQ11D45CLa0KsWROPddfupTDJ/r46ROH2ba3g/ueOsp5S2Zz1coFLG6qD7tEEZHQqaVVoY73DvCLnYe5Z/dR+jI5zmmewYbrzuPyFXNLfq4wEQmd/shHodCqcD39We7dfYSfPXmErr4MF7Y2ceNVZ/Fb555BokznDRORstMf9ygUWhExkM2RTBg3b9nF3qPdtMyp5+XnL+La8xZyYWuTWl8i1UV/0KNQaEXI9ZctJZPNcfvDz3Db/W38fOdhBrLO4tl1XH3eQl5+/iIuXjpHLTCR6NMf8SgUWhEyfPTgsZ4BfvzoQb7/0DNseeIQ/ZkcC2bWcs3qhVx7/kIuXT6XVFKH4olEkEJrFAqtCBlryPuJvgx37mjn9ocO8JPH2ukdyDG3sYarV5/Btect4oqz5pFWgIlEhUJrFAqtCCn2OK3u/gx3PXaI7z/8DHc+epCu/iwzalOcu3gW5y4KLotncXbzDOrSyRJXLSKToNAahUKryg1kc+xsP8HjBzs5cKyXZ4710h9M0pswWDCzlkWz61k0u45Fs+tZOLuOGbVDD9/TQc0iZafQGoUOLq5y6WSC5y6axXMXzQIg587Rrn4OHOvlwLEeDjzby+7DXTyw79mTj5lVl2Lh7DoWzKhl3oxaWubUs3xeI4ub6rSPTERCpZaWANDVlwlaYj1BoPVypKuPgeypjzaVMFrnNrBsXgPL5zUOuW6Z00BNSoEmMk3U0hqFQktG5e509mZ4/tIm9hzp5qkjXSevnzrcRVf/qVnpE5afiqowzJbPa2T5/Hygad+ZyIQotEah7kEZlZkxqz7NZSvmcdmKeUPWuTtHuvrZc6SLpw5356+DQPvvB/bT2ZspeB5YPLueZfMaWDavkeXB9eKm/P6zGbUpZtSlqE8ndZC0iIxJLS2Zdu5OT3+WI139HOnq48iJ/vztE30c6eqnu3/k84YZUJNKUJtKUJtKsripjsbBUKtN0VCbpCaZJJ00UkkjnUwEFyOVSJBOJUgnjFSwbHB9KmmkE8F2yQQ1g8uSp5alg20GX1/77iRk+u9tFGppybQzMxpqUzTUpmid23Da+p7+LEe7+jnWM0BfJktfJkd/Jnfydl8mR99Alu7+LB3dA0PWZXNONufk3MmV8F+VhEFtKkltOh9ytel8kNamEkOCtS6doLEmRWNtKgjYZMHtYctObpekNqXuUpHJUGhJ2dXXJFlSU8+SOVM73UrOnVwQYln3gkCDTC5HLseQ5YNhlx32mFzOyQTrMtn87Uwud+p2NjfkurM3Q0dweyA7GLL54M0W2XORNMuHXzofgEua6ocFXT7cGmtT1IzQ6hvsRrWT9wvWDd/Ghi4v3NiCu6lEvrWaSp66TieNZOJU6zW/Lr++Np2gLp2kLhVcp5MkNX2YlIFCSyIrYUYiaVRSoyWTy9E/kDvVYixoPfYPaUnm1/UH9+trkhzvzY/g7O7LcKIvQ1d/lmwpm5PTbLBFmg+xBHWp5Knb6eSpSypx2vLaVIL6muTJx+S7aE/v4h0M01PBemp90iz/O5Gg4LaCtNootESmUSqRIFWboKF2/G3H455v6Q0PruGNOccL7xReMTzyRtqH7c7JFmjO863TXGHL1J1cjiGt1MGW50A2x0DWT13ncsHy/LL+TI6uvsyQ7TI5J2HQO5CjN5M97f1Mt2QiH2hmQ29PdNDPZPb/T/atPfThqyf5yOpXFQMxzOwHwPyw6xhmPnA47CJGodomR7VNXKXWBZVdW527nxd2EZWoKkKrEpnZVne/JOw6RqLaJke1TVyl1gWqLao0rldERCJDoSUiIpGh0Cqdm8MuYAyqbXJU28RVal2g2iJJ+7RERCQy1NISEZHIUGiJiEhkKLRERCQyFFoiIhIZCi0REYmMqph7sGnlC/z8t/89dekEvQP5uc3e/ZvncOmKuWGXFluD8+b1Z3L0Z3MMBNf9mfwcdCeXB8tOrRt+30ddf/J2xk97rsHXi8p8s+lkftb3mmAC2JrgFCjp4Pxf+duntqlJ5s8fNrju1P38JLK1hY8f/rwjPF86lSARgRNwmhlGfrLk/PyB+WUJAyO4DtYlCrbFOLkuEawjWFe4vIJOQlp0IRdcsdbf909fHnHd9ZctnbaCCu06dIJbt7Xxn/e3cfB4H3Ma0lx34RJee0kLqxfPno6XGPX9V0VozU32Ma+xlmeO97BwVj2vf0FrrAPL3fNf9uMGQsEXfsHyoYEweN+H3T89IE6GTHA7CgxIB+fHGi0g0qcFw8jbnB4WNka4FIaJVdKX5aScFgaJU/eNU0EyJHAwLMHJ9YPBczKMGBo+k5nkNg46nz1antfpHeB7Dx7g1m1tbNvTQcJg7XOa+fArW/jN5zaX7RxxVRFajbUpPvl7F4RdBpCfCfu0AMj40KAYMURGCY/B+yO1UoYHUME2UZAwTv8iH/KlPsqX/pAAsHECobDVUrBtECypRPQDYyJGCodEYctkWItleOuksIUzeL/CWigyjXI55+7dR/jW1ja+//ABegdynLWgkfdfu4pXP38JzbPqyl5TVYTWoGyuoHUxTkCM3M1U+JjRtxmr5ZGJSH9UMmEjfomPer8gIIZ3M53+WBv3uWpSCZ00sAiFQTGk9TJCaIwWLEO70/SZy/j2He3mW9vauO3+Nto6ephZm+LVF7Xw2otbuLC1KdTfo6oIrZ3tJ3jZP26JzAnz0kkb1hVV+GVuI+x7GL9VMeq+j+H7NYJrBUZpjBQyw1suJ7dJFOxToTCI0AkMpex6+rPc/vABbt3axi93HcEMXnjWfP786udw9eqF1KUr42yrVRFagyenK8aYX+gj7agevt1prQYbIXiGPl8Ud3jHSWGYjNaCSQbdiMNbMwoZiTJ35/69Hdy6tY3vPniAE30Zls5t4L0vW8lrLm5hSVN92CWepipCa0lTPR99zfmjtiqqaYe3DGuxDOsyG2zVDG/NFHaRVfCIMZGyOHi8l9vub+Nb29rYdaiLhpokLz9/Ea+9uIUXLJ9b0f+AVUVoNdamuGR5fEcLTtS9u45yy337OHC8h0UhjrY82aJJ5FsyhS2cZGIwkPKnR08E4ZNUi0ZkUvoyWX70yEG+ta2NLY8fIudw6fK53HjVWbz8/EXMqI1GHESjSpk29+46yqfufIJUwphVl+JIVx+fuvMJ3s3ox7VZEBYjDQYYaaTZkNYMp482M1DwiJSBu/Pw08e5dds+/vuB/RzrGWDR7Dr+aO3Z/O7FLSyf3xh2iROm0KpiI+2r+ea2fdQkjfqaFAbMqEvQ05/ltl+18coLF59s+Wg4s0h0nejL8MC+Z/nyL59ixzOd1KQSXL16Ia+9uIUXnj0/0gOxFFoVYvjxMyMdnDkdI8/aO/toqk8PCaIZtcYzx3ppjEj3gIicLptzHj/YybY9Hex45jg5hwtam/jb3zmPV16wmNn16bBLnBb6lpqkYkacFQbM4ECAsAcDtM5poL2zl4aaUz/6noEsLXMayvL6IjK9BrI5vv/QAX6171m6+jLMqE3xwrPmc9GyOfzpy1aGXd60i1VoTeVAzWoZ2rx+zQpu2rSd7v4M9ekkPQNZBrLO+jUrwi5NRCahvbOPXz55hFWLZnLR0jmsPGNmpLv/xlMVoZVMGHMaak4G0fCRaFEPmum0dlUzG4CNW3bR1tFNy5wG1q9ZwdpVzWGXJiKTkE4keN+1qyIz+m+qquJdphLGnMaasMuIjLWrmhVSIlViIJejbyAbm9DS+bRERCLMgC1PHA67jLJRaImIRFhDTYr793ZwvHcg7FLKQqElIhJhM+pS5HLOL3bGo7Wl0BIRibBUwji/ZTb37D5KT3827HJKTqElIhJxa85ZQF8mxz27j4RdSsnFY7hJgc072tm4ZRf7Orpp1XBvEakCi5vqWXnGDH6+8zAvPHs+6WT1tkeq952NYPOOdm7atJ32zl6a6tO0d/Zy06btbN7RHnZpIiJTctXKZrr6s2zb0xF2KSUVq9DauGUX6aTRUJPCLH+dThobt+wKuzQRkSlZPq+BpXMb+OkThyJzFvfJiFVo7evopn7YKaPr00naOrpDqkhEZHqYGVetXEBH9wAPPf1s2OWUTKxCq3VOAz0DQ0fXaLJYEakWz1k4k+aZtdz1+CHcq7O1FavQWr9mBQNZp7s/g3v+WpPFiki1SJixZuUCDh7v47GDnWGXUxKxGj04kcliNcpQRKLogpYm7njkIHc9fijsUkoiVqEFxU0WOzjKMJ20IaMMNwSPFxGpVMmE8aJz5vPdBw9w31NHecHyuWGXNK1i1T1YLI0yFJEou2TZXBpqknxu85NhlzLtFFoj0ChDEYmymlSCK8+ax5072nn0wPGwy5lWVRFauw51TesBwhplKCJRd/mKeTTWJNl4V3W1tqoitDK53LTObKFRhiISdQ01Ka6/bCnfefAA+45WTy9RVYRWwmxa9zmtXdXMhnWraZ5Zx7GeAZpn1rFh3WoNwhCRSFk+v5Fszjl0oi/sUqZN1YwenO59TjolvYhEWc6df//Zbs5bMovntzaFXc60qYqWFmifk4hIoUf2H2fX4S7ecdXZmFnY5UybqgitnLv2OYmIBNydux4/xJnzG7nmvIVhlzOtqqJ7MJVITHmfk2bAEJFq8eShLp5+toePvvp8konqaWVBlbS0VixonHJg6TxbIlIt7nq8nZl1KV510ZKwS5l2VRFaU6UZMESkWrR1dPPkoS5edPZ8alPJ8R8QMQotNAOGiFSPux4/RF06waVVNufgIIUWmgFDRKpDe2cvj+w/zhUr5lGbrr5WFlRJaE11GifNgCEi1eCnTxwmlTSuOGt+2KWUTFWE1lSncdIMGCISdcd6Bnhg77NcvGwuM2qrYmD4iKrinRVO4zTZoNEMGCISZT/feRjHefHZ1dvKgippaYEGTohIfHX3Z7h391Ge19LEnMaasMspqapoaYEGTkyUDqYWqR6/3HWE/myONSsXhF1KyVVFS0vTOE2MDqYWqR7u8Msnj7Bq4UwWzqoLu5ySq4rQmo5pnOJEB1OLVI+u/gzd/VmuikErC6qke7CYaZzUHXbKvo5umurTQ5Zpn6BINJ3ozXDuvAaWzWsMu5SyqIqW1njUHTaUDqYWqR5Zd65aGZ9/wGMRWuoOG0oHU4tUj3QiwcozZoRdRtlURWiNNyOG5hYcSgdTi1SP+ppkVZ3kcTxVsU9rcEaMDTDiF2/rnAbaO3tpqDn1duPeHaaDqUWqQyaXC7uEsqqKllbhjBgjiUJ32OYd7bzh5rt50d/fyRtuvju2+9tEZGL6M/EKrapoacHY3X1rVzWzgfy+rbaObloKRg9WwqjCwYEi6aQNGSgyWstRRGRQJpf/R7ywJ6mahfIuzey1wIeB5wKXuvvWgnUfAN4OZIF3ufsPi3nO8br7RuoOq5SwKBwoAtBQk6K7PzOluRRFJD72He3mOQtnhV1GWYTVPfgw8GpgS+FCMzsXeD2wGrgG+KyZjXtSmMnOiFEpowo1UEREpmLP0fh8V4TS0nL3R4GRRrxcB9zi7n3AbjPbCVwK/HKs5xs+I0axXX6VcpCtBoqIyGSlkwn2HolPaFXaQIwlwL6C+23BstOY2Q1mttXMts6ke0hgFXsgcaUcZBuFgSIiUjkKv/8s209bRw/ZnIddVlmULLTM7A4ze3iEy3VjPWyEZSP+JNz9Zne/xN0vWbDg1JxbE+nyq5Sw0HFTIjIRhd9/jQ0N9GdzHDzeG3ZZZVGy7kF3f+kkHtYGtBbcbwH2T+QJJtLlN9aownLTcVMiMhk1qXzbY8/RbhY31YdcTelV2hjJTcDXzOyTwGLgHODeiTzBRPcPKSxEJMpSCWNGXYq9R7q4YsW8sMspuVD2aZnZq8ysDbgC+J6Z/RDA3bcD3wQeAX4AvNPds6M/0+kqpctPRKRcls5tYG9MRhCGElru/m13b3H3Wnc/w92vLlj3EXc/y92f4+63T/S5tX9IROJm6bxGOroHON47EHYpJVdp3YPTQl1+IhIny+bmd3/sPdLNeUtmh1xNaVXakHcREZmgRU11pBIWiy7Cqgit8U5NIiJSzVKJBEua6hVaUTF4ahIFl4jE1dJ5DTz9bA8D2eqe9b0qQmu8U5OIiFS7ZXMbyOac/c/2hF1KSVVFaIEmmBWReGsdHIxR5V2EVRNammBWROJsZl2aGbUpnjlW3dM5VUVoTfbUJCIi1aK7P8OJvgwLZtaGXUpJVcVxWsNPTSIiEhdzG2u4/rKl/CQYiPa2F53J5VU8nVNVtLRWLGhUYIlIrG3b00EyYTyvRQcXi4hIhdu2p4NzF80aMll4NVJoiYhEXCab44F9z3Lxsjlhl1JyCi0RkYjb8UwnPQNZLlJoiYhIpbt/bweAWloiIlL5tu3pYOGsOhbPrgu7lJKritDShLkiEmfb9nRw0bImzCzsUkquKoaZDE6YuwGmNPR98452Nm7Zxb6OblrnNLB+zQoNpReRijaQdQ539PDWK5eHXUpZVEVLazomzN28o52bNm2nvbOXpvo07Z29mjleRCped38GiMf+LKiS0IKpT5i7ccsu0kmjoSaFWf5aM8eLSKXr7s9Sk0qwenF1H1Q8qGpCa6oT5u7r6KY+nRyyTDPHi0il6+7PcEHLbGpSVfN1PqaqeJfTMWFu65wGegayQ5Zp5ngRqXQ9/VkuWhqPrkGoktCajglz169ZwUDW6e7P4J6/1szxIlLpHGJxUPGgqgit6Zgwd+2qZjasW03zzDqO9QzQPLNOM8eLSCTEqaVVFUPep8vaVc0KKRGJlJpkourPoVWoKlpaIiJx1VCbHH+jKlIVoaUZMUQkrqr9VCTDVUVoDc6IoeASkbhx97BLKKuqCK3pmBFDRCSK+jO5sEsoq6oILdCBwCISTwqtiNKBwCISR/1ZhdZJltdUcD9tZm8zs4dLX1rxpmNGDBGRKOrP5Mjl4rNfa9TQMrPXAh3A42Z2p5mtAXYCrwLeVqb6ijIdM2KIiESRAwc7e8Muo2zGGiv5V8Bl7v6Ymb0A+BnwRne/tTylFW86ZsQQEYmqvUe6WTS7PuwyymKs7sF+d38MwN3vA56qxMASEYm7PUfjMwhtrJZWs5m9q+B+Y+F9d/906coSEZFi7VNoAfD/gAVj3BcRkZDVJBPsVWiBu3+onIWIiMjE1aQUWgCY2SeHLXLgMPATd7+7pFWJiEhRalIJdQ8Gto+wbC7waTP7irv/c4lqEhGRItUkExw+0U9XX4bG2uqfPHes7sEvjLTczD4D/AJQaImIhKwmlWAA2NfRzaqFs8Iup+QmPI2Tu3eT7yoUEZGQ1STzX+N7jsSji3BCoWVmCTN7E7C/RPWIiMgE1KTyX+Nx2a811kCMDk5vUWWBnwLrS1mUiIgUJ5kwZtWlYjOCcKy9dvOH3Xd3j9d0wiIiEbB0XkNsQmus7sHXuXvW3bPACwoDy8zeUfrSRESkGEvnKrQA/rzg9meHrftfJahFREQmYSDrpBIWdhllMVZo2Si3R7ovIiIh2Xe0m6VzG8MuoyzGCi0f5fZI90VEJCR7j3azdG48ztw+1kCMVWZ2P/lW1XOC2wT3V5a8sgnYdaiLzTvadU4tEYmdTM7p7s+ydG48zqc1VmidX7YqpiiTy3HTpu1sAAWXiMRKfyY/Rm7pvJi3tNz9yXIWMhUJM9JJY+OWXQotEYmVk6EVk+7BCU/jVKnq00naOuIx5FNEZFB/Nh9aLXMUWpHSM5CNzQ9NRGRQfybHGbNqqUsnwy6lLKoitHLuDGSd9WtWhF2KiEhZ9WdysekahEmElpl9wcz+2cxWlaKgyUglEmxYt1r7s0QkdvqzudgcowWTa2l9HvgZU5gVw8w+ZmY7zOxBM/u2mTUVrPuAme00s8fM7Opinm/FgkYFlojE0kBWLa0RmVktgLvf7e7fcPc/m8Lr/gg4z92fBzwOfCB4jXOB1wOrgWuAz5pZPDpqRUQmaem8eByjBUWElpldamYPAU8E9y8wsymdtdjd/8fdM8Hdu4GW4PZ1wC3u3ufuu4GdwKVTeS0RkWqnltZQnwZeARwBcPdfA78xjTW8Dbg9uL0E2Fewri1Ydhozu8HMtprZ1kOHDk1jOSIila3w+w+gVaE1dBt33zNsWXa8B5nZHWb28AiX6wq2+SCQAb46uGiEpxpxnkN3v9ndL3H3SxYsWFDE2xARqQ6F339msGBGbdgllc1Y0zgN2mdmlwIe7F/6E/L7ocbk7i8da72ZvYV8C+4l7j4YTG1Aa8FmLcD+ImoUEYmlmmQCs/iceKOYltY7gPcCS4GDwOXBskkzs2uA9wHr3L1wGotNwOvNrNbMzgTOAe6dymuJiFSz2lRVHG5btHFbWu7eTn5E33T6F6AW+FHwH8Ld7n6ju283s28Cj5DvNnxncOZkEREZQU0qXgOsxw0tM/vkCIuPAVvd/XuTeVF3P3uMdR8BPjKZ5xURiZuaZHy6BqG47sGZwGXkR/XtA14ALAT+yMw+UcLaRERkHIlEvEKrmIEYZwFr3X0AwMz+BfgBcDXwa2AqBxmLiMgUZLLxOpF8MS2tJUDh4db1wJLg4OC+klQlIiJFyeTiFVrFtLQ+CTxgZj8mfxzVWuBjZtYIbC5daSIiMp5McD6tuChm9OBGM/se+f1aBvy1uw/OWvHeUhYnIiJji1tLq9gB/p3AbvIH+raa2ZWlK0lERIo1oJbWUGb2NvKDLZYAD5EfPXg3+W5CEREJkVpap/tT4BLgKXd/MXAxcKCkVYmISFGyOY9Va6uY0Op19x4AM6tx9+1AxZy1WEQk7o529YddQtkUM3rwQHBm4e8APzSzo+TnIBQRkQpwqLOPM2bVhV1GWRQzenBdcPNDZvYSYDbw3ZJWJSIiRTt0Ij6HzBZz5uK3DN529x+7+38CN5W0KhERKdrhzviEVjHdg9ebWZ+73wJgZp8GZpW2LBERKdbhE9qnVehVwHfMLAtcC3S7+1tLWpWIiBQlYcZhdQ+Cmc0ys1nkg+0twF8C3cBfBstFRCRkqYRxSN2DAGwHnPzUTYPX1wUXJ38mYxERCVEqGa+W1qih5e6t5SxEREQmLpVIKLSGM7NLgeWF27v710pUk4iIFCnf0tJAjJPM7IvAuTU1cKYAABnRSURBVMADQDZY7IBCS0QkZKmEcbSrn4FsjnSy2DnQo6uYltblwLnuHp/JrUREIiKRMAB6BrKxCK1i3uF2YH6pCxERkYnLZp2aZIKZtUXt7Ym8Yt7lbOBRM7sbOLm3z91fXbKqSmjzjnY2btnFvo5uWuc0sH7NCtauag67LBGRScnknHkzajCzsEspi2JC6+9KXsUU7TrUxeYd7eOGz+Yd7dy0aTvppNFUn6a9s5ebNm1nAyi4RCSSMtkcC2bWhl1G2RQzYe6Py1HIVGRyuaLCZ+OWXaSTRkNN/m031KTo7s+wccsuhZaIRNJAzpk/Iz6hNdaMGHcF1x1mdrTg0hGcnqRiJMxIJ42NW3aNud2+jm7q08khy+rTSdo6uktZnohIyWSyOebPqAm7jLIZq6X1G8F1JAZhFBM+rXMaaO/sPdnSgvyIm5Y5DaUuT0SkJLJqaeW5e87MksD97p4dfiljjUUpJnzWr1nBQNbp7s/gnr8eyDrr16woU5UiItPLIVb7tMYc8h6E0yNmtqRM9UxKzr2o8Fm7qpkN61bTPLOOYz0DNM+sY8O61dqfJSKRFqeWVjGjB+eTH/L+S6BrcGElDXlPJRJFh8/aVc0KKRGpKgqtoT5a8iqmaMWCRgWRiMRWnLoHq2LIu4hInC2IUUtr3GmczOw6M3vUzI6Z2XEz6zSz4+UoTkRExmbArPp4TOEExc09+Angde4+291nuftMd6+oMxc/sv84n77j8bDLEBEpu1QyEZspnKC40Dro7g+VvJIpyLnzqTt3KrhEJHZSifgEFoyxT8vM1gU37zOzrwL/xdAJczeVuLaimRkJg3/72W7e9dKVYZcjIlI2qaRCa9BrC27ngHUF9x2omNACSBh09VfcMc8iIiWVSlT/ObQKjRpa7v6mchYyVTmHxprk+BuKiFSRuLW0ihk9+HdmNsvMUmb2QzM7aGbXl6O4Yrk7OYc/fNGZYZciIlJWcdunVUy78lp3Pw68AmgHzgPeV9KqJihhxrt/82ztzxKR2FFonW6wC/HlwNfd/RD5fVoV49zFsxRYIhJLuYr6Ni69Yo5Iu93MHgaywDvNbD4FowhFRCQ82Zil1rgtLXf/c+A3gYvdfQDoBSpmslwRkTiLW2gVNfeHu7cX3D4BnChZRSIiUrSsxyu04jXAX0SkysStpaXQEhGJMIXWMGZ2i5ldbXGakVFEJCIUWqf7IvA24HEz+1szO7u0JU3crkNdbN7RPv6GIiJVRqE1jLv/wN1/D7gUeAb4iZltMbM3mVlFnMQlk8tx06btCi4RiZ24DcQoKnTMbA5wPfAm4EHga8CLgLcALy1ZdUVKmJFOGhu37GLtquYJPXbzjnY2btnFvo5uWuc0sH7Nigk/h4hIWLI5x91jc06tYvZpfRP4BTAXeI27/7a7f9Xd3wHMK3WBxapPJ2nr6J7QYzbvaOemTdtp7+ylqT5Ne2evWmwiEjlxOsNFMfu0/g04193/xt3bCle4+/NLU9bE9QxkaZnTMKHHbNyyi3TSaKhJYZa/HmyxiYhExbGegbBLKJtiTgIJ8MrhTc9KOglkzp2BrLN+zYoJPW5fRzdN9ekhyybTYhMRCdOx7gGWNNWHXUZZFHsSyOEq6iSQqUSCDetWT3hfVOucBto7e2moOfUxTKbFJiISpuO9amlF6iSQKxY0TmrwxPo1K7hp03a6+zPUp5P0DGQn1WITEQmTugcBM3uDu3/dzN410np3/3TpyiqPtaua2UB+31ZbRzctGj0oIhGk0MqbE1wvmO4XNbO/Aa4DcuRPLPlWd98fzLrxKfLn7uoOlt8/3a9faO2qZoWUiETacYUWuPtng+sPleB1Pzb4vEFL7ibgRuBa4JzgchnwueBaRERGodAqYGa1wFuB1UDd4HJ3v2GyL+ruxwvuNnLqTMjXAV92dwfuNrMmM1vk7gcm+1oiItUsYcbx3kzYZZRNMcdpfRlYDrwCuAc4i/yJIKfEzD5iZvuA3yff0gJYAuwr2KwtWDbS428ws61mtvXQoUNTLUdEJDIKv/9y7tSlk2GXVDbFhNZKd/8AcMLdvwBcA5w33oPM7A4ze3iEy3UA7v5Bd28Fvgr88eDDRniqESfWcveb3f0Sd79kwYJp3+0mIlKxCr//AGbVV8Q0sGVRzDsd7Cx91syeCxwElo33IHcvdk7CrwHfA/6KfMuqtWBdC7C/yOcREYml2cMmSahmxbS0vhBMmPtXwA+Bx4FPTuVFzeycgrvrgB3B7U3Amy3vcuCY9meJiIwtTqE1bkvL3TcGN38CLJ2m1/2omT2H/JD3PeRHDgJ8n/xw953kh7z/wTS9nohI1ZpVp9BitIOKB03l4GJ3f80oyx1450Sfb/AkkDreSkTiKE4trbG6BxcElyuB95AfNXgW8G7gwtKXVjydBFJE4mxWjEJrrIOLBw/+/SFw4eCxVWb2IeAb5SmvOFM5CaSISNSppTXUMoYel9UHnFmaciZPpxQRkbiaVach74W+BtxjZreRP2bq1cB/lLSqSdApRUQkjhJmpJLFtD+qQzGjBzeY2e3AmmDRje5+X2nLmpjJngRSRCTqkomR5mSoXkW1KYOQqqigKjTZk0CKiERd0uIVWlXRppzsSSBFRKIubi2tqggtEZG4UmiJiEhkxC20xpoRo4ORZ1g38pNXzC1ZVSIiUhSF1inzy1aFiIhMStwGYow1I0a28L6ZzaXgzMXolCEiIqFLxGwnz7hv18x+28weJ3+uq3uC6ztLXZiIiIwvbt2DxWT0R4AXAo8FZxq+GthcyqJERERGUkxoZdz9EJAwM3P3HwEXlbguEREpQm6k4XJVrJjQOmZmjcDPgC+b2SfIn7yxYgyeT0tEJG5yMUutYkLrd8jP8v4e8t2CTwOvKGFNE6bzaYlIXOVcoTXcB9w96+4D7v4Fd/8k8N5SFzYRhefTEhGJk6xC6zTXjLDst6e7kKnS+bREJI5yFbWzpvTGmhFjPXAjsNLM7i9YNRPYWurCJkrn0xKROIpb9+BYM2J8E/gx8HfA+wuWd7p7Re080vm0RCSuFFoBd+8AOoDXmtl5wIuCVT8FKiq0dD4tEYmruHUPFjMjxjvJt7qWBpdvmtkflbqwidD5tEQkrtTSOt164FJ3PwFgZv8X+AXw2VIWJiIi44tbaBUzetCAgYL7A8EyEREJWTZmBxePNXow5e4Z4CvA3WZ2W7DqVcCXylGcTMzmHe1s3LKLfR3dtM5pYP2aFeo2FalyMcusMVta9wK4+z8ANwDdQA9wo7t/vAy1yQRs3tHOTZu2097ZS1N9mvbOXs0SIhIDOXc8Rl2EY+3TOtkF6O73AfeVvhyZrI1bdpFOGg01+R9pQ02K7v4MG7fsUmtLpMr1DGRP/u1Xu7He5QIzG3W6pmA6J6kQ+zq6aapPD1mmWUJE4qGrT6EFkARmoEEXkdA6p4H2zt4hv7iaJUQkHrr7M0Bt2GWUxVihdcDdN5StEpmS9WtWcNOm7XT3Z6hPJ+kZyGqWEJGYONGXCbuEshlrIIZaWBGydlUzG9atpnlmHcd6BmieWadZQkRiors/G3YJZTNWS+slZatCpsXaVc0KKZEYitOJIEdtabn70XIWIiIik9NYG49BGFDcjBgiIlLBFFoiIhIZjTXJsEsoG4WWiEjENailJSIiUdGQVktLREQiIGFGIhGfI5QUWiIiERajvAIUWiIikZaweKWWQktEJMLi1DUICi0RkUhLxiuzFFoiIlGm7kEREYkMdQ+KiEhkqKUVQbsOdbF5R3vYZYiIlF3MGlrVEVqZXI6bNm1XcIlI7Kh7MIISZqSTxsYtu8IuRUSkrNQ9GFH16SRtHd1hlyEiUlZJhVY09QxkaZnTEHYZIiJllaiab/HiVMXbzbkzkHXWr1kRdikiImVlqKUVOZms87sXLWHtquawSxERKSt3D7uEsqqK0EoljW/d/7RGD4pI7OTilVnVEVoaPSgicZVTS6t8zOx/m5mb2fzgvpnZp81sp5k9aGYXFftcGj0oInGk0CoTM2sFXgbsLVh8LXBOcLkB+Fyxz6fRgyISRzHLrFBbWv8I/AVQ+JFfB3zZ8+4Gmsxs0XhPpNGDIhJXammVgZmtA552918PW7UE2Fdwvy1YNtJz3GBmW81sa7b7OBvWrdboQRGJhcLvv+6e3rDLKatUqZ7YzO4AFo6w6oPAXwK/NdLDRlg24r8R7n4zcDPAJZdc4gosEYmLwu+/M1acG6umVslCy91fOtJyMzsfOBP4teWnH2kB7jezS8m3rFoLNm8B9peqRhGRqNOQ9xJz94fcvdndl7v7cvJBdZG7PwNsAt4cjCK8HDjm7gfKXaOISFTE7eDikrW0Jun7wMuBnUA38AfhliMiUtni1tIKPbSC1tbgbQfeGV41IiLRotGDIiISGTHLLIWWiEiUqaUlIiKRodASEZHIiFlmKbRERKJMLS0REYkMB7IxGveu0BIRibiBbC7sEspGoSUiEnE1yfh8lcfnnYqIVCEDEomR5hqvTgotEZEIS1h8AgsUWiIikRazzFJoiYhEmVpaIiISGQotERGJjJhllkJLRCTK1NKKoF2Huti8oz3sMkREyi5Go92BKgmtTC7HTZu2K7hEJHZMLa3oSZiRThobt+wKuxQRkbJKVMW3ePGq5u3Wp5O0dXSHXYaISFlpn1ZE9QxkaZnTEHYZIiJlFbPMqo7QyrkzkHXWr1kRdikiImWlllYEpRIJNqxbzdpVzWGXIiJSVgkUWpGzYkGjAktEYsmq4lu8eDF7uyIi1UXdgyIiEhlx+xKP2/sVEakqOrhYREQixMMuoKwUWiIiEZaLV2YptEREoswVWiIiEhWu7kEREYkKdQ+KiEhkeMz6BxVaIiIRFrPMUmiJiERZLmappdASEYmweEWWQktEJNJi1tBSaImIRJm6B0VEJDJillkKLRGRKNOQdxERiYx4RZZCS0Qk0rRPS0REIiNmmaXQEhGJMrW0REQkMmKWWQotEZEoi1lmKbRERCQ6FFoiIhGWsLArKC+FlohIhBnxSi2FlohIhFm8MkuhJSISZYmYpZZCS0QkwmKWWQotEZEoU2iJiEhkqHtQREQiI16RpdASEYk0U0ur9Mzsw2b2tJk9EFxeXrDuA2a208weM7Orw6hPRCQq4nZwcSrE1/5Hd/944QIzOxd4PbAaWAzcYWYr3T0bRoEiIpVOLa1wXQfc4u597r4b2AlcGnJNIiIVK2aZFWpL64/N7M3AVuDP3L0DWALcXbBNW7DsNGZ2A3BDcPeEmT1WymInYT5wOOwiRqHaJke1TVyl1gWVXdvD7n7eaCuHff9tB0bdttqYl+hkLGZ2B7BwhFUfJB9Mh8nPqv83wCJ3f5uZfQb4pbv/R/AcXwC+7+63laTIEjKzre5+Sdh1jES1TY5qm7hKrQtUW1SVrKXl7i8tZjsz+zzw3eBuG9BasLoF2D/NpYmISESFNXpwUcHdVwEPB7c3Aa83s1ozOxM4B7i33PWJiEhlCmuf1j+Y2YXkuwefAtYDuPt2M/sm8AiQAd4Z4ZGDN4ddwBhU2+Sotomr1LpAtUVSyfZpiYiITLdKG/IuIiIyKoWWiIhEhkJrGphZq5n9xMweNbPtZvbuYPmo01WVub6nzOyhoIatwbK5ZvYjM3siuJ5T5pqeU/C5PGBmx83sPWF9Zmb272bWbmYPFywb8TOyvE8H0409aGYXhVDbx8xsR/D63zazpmD5cjPrKfj8/jWE2ipimrZRavtGQV1PmdkDwfKyfW5jfF9UxO9bxXN3XaZ4ARYBFwW3ZwKPA+cCHwb+dwXU9xQwf9iyfwDeH9x+P/D3IdaXBJ4BloX1mQFrgIvIH9Q55mcEvBy4nfwE25cD94RQ228BqeD23xfUtrxwu5A+txF/hsHfxK+BWuBM4EkgWc7ahq3/BHBTuT+3Mb4vKuL3rdIvamlNA3c/4O73B7c7gUcZZSaPCnId8KXg9peA3wmxlpcAT7r7nrAKcPctwNFhi0f7jK4Dvux5dwNNww7jKHlt7v4/7p4J7t5N/pjGshvlcxtNWadpG6s2y0/Y9zrg66V6/dGM8X1REb9vlU6hNc3MbDnwfOCeYNEfB036fy93F1wBB/7HzLZZfvoXgDPc/QDk/4iA5pBqg/wkyYVfHpXwmcHon9ESYF/BdqNON1YmbyP/n/igM83sV2Z2l5m9OKSaRvoZVtLn9mLgoLs/UbCs7J/bsO+LqPy+hUqhNY3MbAZwG/Aedz8OfA44C7gQOEC+OyIML3T3i4BrgXea2ZqQ6jiNmdUA64Bbg0WV8pmNZaQpSkM5dsTMPkj+mMavBosOAEvd/fnAe4GvmdmsMpc12s+wYj434A0M/Uep7J/bCN8Xo246wrLYHquk0JomZpYm/wv4VXf/TwB3P+juWXfPAZ8npBnr3X1/cN0OfDuo4+BgF0Nw3R5GbeSD9H53PxjUWBGfWWC0z6giphszs7cArwB+34OdH0HX25Hg9jby+41WlrOuMX6GlfK5pYBXA98YXFbuz22k7wsq/PetUii0pkHQP/4F4FF3/2TB8tGmqypnbY1mNnPwNvkd+A+TnzLrLcFmbwH+u9y1BYb8x1sJn1mB0T6jTcCbg1FdlwPHBrt1ysXMrgHeB6xz9+6C5QvMLBncXkF+KrRdZa6t0qdpeymww93bBheU83Mb7fuCCv59qyhhjwSphgvwIvLN9QeBB4LLy4GvAA8FyzeRn82+3LWtID9i69fkT2HwwWD5PODHwBPB9dwQamsAjgCzC5aF8pmRD84DwAD5/2zfPtpnRL675jPk/xt/CLgkhNp2kt/PMfj79q/Btq8Jfs6/Bu4HXhlCbaP+DMmf5eFJ4DHg2nLXFiz/InDjsG3L9rmN8X1REb9vlX7RNE4iIhIZ6h4UEZHIUGiJiEhkKLRERCQyFFoiIhIZCi0REYkMhZaIiESGQktERCJDoSUiIpGh0BIRkchQaImISGQotEREJDIUWiIiEhkKLRERiQyFloiIRIZCS0REIkOhJSIikaHQEhGRyFBoiYhIZCi0REQkMhRaIiISGQotERGJDIWWVCQzqzMzN7OWIrf/qJn9W5HbftHM/iK4fY2Z7ZxKrRMx0fclIkMptKRoZnai4JIzs56C+78/zmPLGg5jcfe3uvs/hF1HqQSfdW7Yz+uEmT0/WP+Mmb1o2GNuNLM7gturgmAdfNwuM3tvGO9FZLhU2AVIdLj7jMHbZvYU8Ifufkd4FckYdrn72VN4fHbw521mVwB3mtl97v7T6SlPZHLU0pJpY2b1ZvYZMztgZm1m9jEzS5vZPODbwIqC/97nmdkLzeweMztmZvvN7B/NrKh/pMzsbDP7uZl1mtntwJyCdSkzu83MDprZs2b2EzN7TsH6W8zs/4zwnB8ys68OW/Z5M/voKDV8KHivx83sUTN7cbC86PcVfGb/ZGb7ghbQP5tZbbBuoZn9IHgPR8zszmI+m+nm7r8EngAuDOP1RQoptGQ6/TXwPOB84GJgLfAX7n4EeBX5//5nBJcjwADwx8Bc4MXAK4E/LPK1vglsAeYBHwfeNGz9JuAsYCGwA/hSEc/5ZWCdmQ22MGqB3wW+MnxDM7sA+APyX+Szgd8G2oLVE3lf/wi0kP/MngOsBN4frHsf8BgwH1gEfLiI9zCtLO/FQV0V0b0r8abQkun0+8Bfufthdz8I/C2nh8lJ7n6vu9/n7ll3fxL4N+Cq8V7EzFYC5wJ/7e797v5j4AcFz5tx9y+5+wl37yUfppeaWd1Yz+vue4Ct5AMW8mGz2923j7B5BqgP6ki6+y533z2R9xW0vt4GvNvdn3X3Y8BHgdcHmwwAi4GlwfvcMt5nU+DMoIVWeElO4PFJM3sW6Cb/z8En3P32CTxepCQUWjItzMzIt2r2FCzeAywZ4zHnmtntQTfeceAm8q2K8SwGDgWBVPhag8+bMrOPBwMIjpNvaRn5Vtl4vgS8Mbj9RkZoZQEEQfZ+4CNAu5l91czOmOD7Wgykge2DwQL8F9AcrP8IsB/4iZntnOBgiN3u3jTskg3WZYLXLZQmH5KDsu7eBMwEPgj8RrFdtyKlpNCSaeHuDjwDLCtYvBR4enCTER72eeB+4Cx3nwVsIB8u4zkAzB/WclpacPsPgN8CfoN8192qYHkxz/0t4HIzWx08x9dH2zBozV0JrADqyLcsofj3dYB8gJxVECyz3X1e8PzH3P3d7r4MeA3wf8zshUW8h/HsBZYPW3YmQ//hIKghA/wdUEPxXbciJaPQkun0deCvgkEWzeT/Q/+PYN1BoHlwf1FgJnDM3U8EIfG/inydx8m3nj5kZjVm9hvANcOetxc4AjRyKkzG5e4nyO8P+zqw2d2fGWm7oDV1VbDfqye4DLZkinpf7j4A/DvwKTObH+w/ajWzlwWvsc7MzgxasceC588G624xs38t9n0N8w3gz8zsnOA1LwfeHCwfqU4n3235ATMb3kITKSuFlkynm4BHgO3AA8DPgcHjoX5NPgz2BF1hc4E/Bf7QzE4An2GUL83hgi/R3yPfkjoK/AWnwhHgC8Ah8i2/h4CfTfB9fIn8wIgRuwYD9cAngMPkW0wzyL9/mNj7eg/5LsCt5IPpB8DgUPXnApuBTvL7lT7u7ncH61rJf76jKRypOXh5RbDuM8AtwWsdI7/P7b3u/pMxnu8/yXcfvnWMbURKzvJ//yIyKBjosRVY6O7dYdcznJk1APcBzyvYTyUSCwotkQLBCLvPADl3/6Ow6xGRoTQaSCQQdFnuBXYBV4dcjoiMQC0tERGJDA3EEBGRyFBoiYhIZCi0REQkMhRaIiISGQotERGJjP8Py0/VJy7qjUYAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x432 with 3 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "sns.jointplot(x=\"daily_sales_in_euros\" , y=\"daily_shrinkage_in_euros\",kind=\"reg\", data=df.sample(1000));\n",
    "plt.ylim([-50,-1]) \n",
    "plt.xlim([1,200])\n",
    "plt.xlabel(\"Total daily sales, EUR\", fontsize = 12, labelpad=20);\n",
    "plt.ylabel(\"Total daily shrinkage, EUR\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Figure 12. Visualisation of relationship between sales and shrinkage*"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The histogram on the top of Figure 12 shows the distribution of the total sales. Distribution is positively skewed (mean = 49 is greater than the median = 21).\n",
    "\n",
    "As the last step of the report, let's find what theoretical distribution is the most suitable to describe our sales data? First, let's test sales distributions with a python method and find which distribution fits."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 47,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.7/site-packages/scipy/stats/_distn_infrastructure.py:2381: RuntimeWarning: invalid value encountered in double_scalars\n",
      "  Lhat = muhat - Shat*mu\n",
      "/opt/anaconda3/lib/python3.7/site-packages/scipy/stats/_continuous_distns.py:515: RuntimeWarning: invalid value encountered in sqrt\n",
      "  sk = 2*(b-a)*np.sqrt(a + b + 1) / (a + b + 2) / np.sqrt(a*b)\n",
      "/opt/anaconda3/lib/python3.7/site-packages/scipy/optimize/minpack.py:162: RuntimeWarning: The iteration is not making good progress, as measured by the \n",
      "  improvement from the last ten iterations.\n",
      "  warnings.warn(msg, RuntimeWarning)\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Best fit for sales variable reached using lognorm\n"
     ]
    }
   ],
   "source": [
    "data = df[\"daily_sales_in_euros\"]\n",
    "distributions = [st.alpha,st.beta,st.gamma,st.laplace,st.lognorm,st.pearson3,st.norm]\n",
    "mles = []\n",
    "for distribution in distributions:\n",
    "    pars = distribution.fit(data)\n",
    "    mle = distribution.nnlf(pars, data)\n",
    "    mles.append(mle)\n",
    "results = [(distribution.name, mle) for distribution, mle in zip(distributions, mles)]\n",
    "best_fit = sorted(zip(distributions, mles), key=lambda d: d[1])[0]\n",
    "print(\"Best fit for sales variable reached using {}\".format(best_fit[0].name, best_fit[1]))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Logarithm of the sales variables is normally distributed (Figure 13). We have skewed distribution with low mean values, large variance, and all-positive values "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 48,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/opt/anaconda3/lib/python3.7/site-packages/pandas/core/series.py:853: RuntimeWarning: divide by zero encountered in log10\n",
      "  result = getattr(ufunc, method)(*inputs, **kwargs)\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 0, \"log10('daily_sales_in_euros')\")"
      ]
     },
     "execution_count": 48,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,5))\n",
    "x = np.log10(df[\"daily_sales_in_euros\"])\n",
    "x = x[np.isfinite(x)]\n",
    "ax = sns.distplot(x, kde=True)\n",
    "plt.xlabel(\"log10('daily_sales_in_euros')\",fontsize = 12, labelpad=20)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Figure 13. Normal distribution of logarithm of the sales variables ."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now, we can plot lognormal distribution of sales."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 49,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(0, 120)"
      ]
     },
     "execution_count": 49,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.figure(figsize=(10,5))\n",
    "ax = sns.distplot(df[\"daily_sales_in_euros\"], fit=st.lognorm, kde=False, bins=500)\n",
    "plt.xlabel(\"Total daily sales, EUR\", fontsize = 12, labelpad=20);\n",
    "plt.xlim(0, 120)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "*Figure 14. Visualizing sales lognorm distribution.*"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 50,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "54.0"
      ]
     },
     "execution_count": 50,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x = df[df[\"daily_sales_in_euros\"] > 0]\n",
    "np.quantile(x[\"daily_sales_in_euros\"], 0.75)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "As a conclusion, during the last step, we explored shrinkage and sales relation and could not find any correlation. Sales in euros have lognormal distribution, we proved it by plotting normal distribution of logarithm. 75% of the sales are between 1 and 54 euros."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conclusion\n",
    "By the current exploratory process, we answered all questions that were asked at the beginning of the report and found following:\n",
    "- sales drops on Fridays, while shrinkage raises to 42% out of total on Fridays,\n",
    "- fresh and ultrafresh products take 56% and 44% of total sales and 73,6% and 26,4% of total shrinkage respectively,\n",
    "- the top 3 products with the highest shrinkage take around 7% out of the total shrinkage and make only 0,58% of total sales,\n",
    "- there is no correlation between the sales and shrinkage,\n",
    "- the most suitable theoretical distribution of the sales is positively skewed lognormal distribution and 75% of the sales values lay in the interval between 1 and 54 euros.\n",
    "\n",
    "Also, we made a good ground for further exploration of a longer period of time and tackling problems of shrinkage. Since shrinkage is a very complex and unstable variable, it needs more data from the company and work on site."
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}