main.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "13IjaRwElopp"
   },
   "source": [
    "---\n",
    "# Homework 2 - eCommerce analytics\n",
    "---"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "zQXKIvXKlopr"
   },
   "source": [
    "# Notes before starting"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "KRc7kDaClops"
   },
   "source": [
    "We dealt with a huge amount of data from the files `2019-Oct.csv` and `2019-Nov.csv`. These files have a total size equal to 13.4GB. For all the questions, we reduced the amount of RAM required by loading only the needed columns and, since some columns data allows it, we forced some data casting to data types smaller in size. Specifically, for example, we loaded the `user_id` values as `numpy.int32` instead of default `numpy.int64`, since we observed that `user_id` values are of 10<sup>8</sup> order.\n",
    "\n",
    "Moreover, we tried to handle these data applying two different strategies. The first one was executing the functions in Google Colab. According to the second one, we implemented some functions chunking the dataset, computing by chunk and then combining the results. We are aware that this implementation adds a lot of overhead, but this way we managed to run these functions on a laptop with only 8GB of RAM. In addition, we had the opportunity to learn how to work with chunks. For this sake, we decided to keep these functions inside the main script file `./scripts/our_functions`. All the questions marked with a **\\*** were solved with such functions."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "rmIrdmIslopt"
   },
   "source": [
    "# Setup"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "El_Mvlv4lopt"
   },
   "source": [
    "As previously mentioned, in order to keep the notebook tidy, for each question, we put the functions in the file `./scripts/our_functions.py`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "id": "sQ5Q3xSylopu"
   },
   "outputs": [],
   "source": [
    "import scripts.our_functions as f"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "i3Zhm3Kslopy"
   },
   "source": [
    "Specifically, the libraries we used are the following:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "J8ucHbX2lopy"
   },
   "source": [
    "```python\n",
    "# for data handling and analysis\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "from matplotlib import pyplot as plt \n",
    "import datetime\n",
    "\n",
    "# utilies for computational performance evaluation and debugging\n",
    "from tqdm import tqdm\n",
    "from threading import Thread\n",
    "from queue import Queue\n",
    "import os, glob\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "bzaqyOdUlopz"
   },
   "source": [
    "To have an insight into the data handling, we show the routine function to load the needed data for each question (only for the questions marked with an asterisk): "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "7n9bS4u0lop0"
   },
   "source": [
    "```python\n",
    "def get_needed_iterator(month_files:list, columns:list, chunksize:int=10**6, parse_dates=False):\n",
    "    global DATA_PATH, dtype\n",
    "    iterators = []\n",
    "    date_parser = pd.to_datetime if parse_dates else None\n",
    "    for month in month_files:\n",
    "        iterators.append(pd.read_csv(os.path.join(DATA_PATH, month), usecols=columns, \n",
    "                                     parse_dates=parse_dates, date_parser=date_parser, \n",
    "                                     chunksize=chunksize, dtype=dtype))\n",
    "    return iterators\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "XaJPjBITlop0"
   },
   "source": [
    "According to the pandas API, we split the dataset into chunks of 10<sup>6</sup> rows and we obtain an iterator. Using this iterator properly, we get all the data required for each question."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "egM97M7xlop1"
   },
   "source": [
    "---\n",
    "# [RQ1]  Which is the rate of complete funnels? *"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "QXDKwtevlop2"
   },
   "source": [
    "According to the question statement, a complete funnel involves three steps: 1) view, 2) cart and 3) purchase. Because an event of type *cart* implies (and corresponds to) an event of type *view*, we only select the rows related to a *cart* and *purchase* events. Then, regarding to the *purchase* event, we make a distinction between the purchases after a corresponding *cart* event and the one-click purchases, which don't correspond to a *cart* event. Besides the complete funnels, we consider incomplete funnels the route {1} (orphan views) and the route {1,2} (orphan carts).\n",
    "\n",
    "To be clear, the sets that we consider in order to compute the rate of complete funnels are:\n",
    "\n",
    "$\n",
    "U = \\{\\text{all tuples}\\}\\\\\n",
    "P = \\{\\text{purchase tuples}\\}\\\\\n",
    "C = \\{\\text{cart tuples}\\}\n",
    "$\n",
    "\n",
    "The set $P - (P \\cap C)$ contains only the one-click purchases. So, given the cardinalities\n",
    "\n",
    "$\n",
    "|U| = n_{tuples}\\\\\n",
    "|P| = n_{purchases}\\\\\n",
    "|C| = n_{carts}\\\\\n",
    "|P-(P \\cap C)| = n_{one-click}\n",
    "$\n",
    "\n",
    "we compute the rate of complete funnels without one-click purchases as\n",
    "\n",
    "$\n",
    "r_{\\%}=\\frac{n_{purchases}-n_{one-click}}{n_{tuples}-2\\dot(n_{purchases}-n_{one-click})-[n_{carts}-(n_{purchases}-n_{one-click})]-n_{one-click}} = \\frac{n_{purchases}-n_{one-click}}{n_{tuples}-n_{purchases}-n_{carts}}\n",
    "$\n",
    "\n",
    "Similarly, we compute the rate of complete funnels with one-click purchases."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "xmUoAmMrlop2"
   },
   "source": [
    "We show two results, one that corresponds only to the route {1,2,3} and the other that takes into account also the route {1,3} (one-click purchases)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "JoT08FA7lop3",
    "outputId": "fc46c5ab-749a-43ab-8e8d-83d14b803395"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 1659788 purchases,\n",
      "of which 556992 one_click purchases (only view-purchase couples).\n",
      "The number of carts is 3955446.\n",
      "\n",
      "\n",
      "Rate of complete funnels without one_click purchases:\t1.06%.\n",
      "\n",
      "Rate of complete funnels with one_click purchases:\t1.59%.\n"
     ]
    }
   ],
   "source": [
    "f.complete_funnels(['event_type', 'user_session', 'product_id'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "fdAd3ppKlop6"
   },
   "source": [
    "The results show that the **1.06%** of the funnels, excluding the view-purchase ones, are complete. The interesting outcome is that the rate of complete funnels increases of about 50% if taking into account also the one-click purchases (**1.59%**). We deduce that customers often enjoy the one-click route instead of the \"traditional\" route, maybe thanks to the immediacy of the transaction. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "B7wAtFlPlop7"
   },
   "source": [
    "## What’s the operation users repeat more on average within a session? *"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "1im_peqilop7"
   },
   "source": [
    "In order to get, for each session and event (*view*, *cart*, *purchase*), the average number of operations performed by users, we need to count the tuples of each event and the number of distinct sessions; then, the rate $\\frac{n_{event}}{n_{sessions}}$ is the requested outcome."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 564
    },
    "id": "cgft8aKXlop8",
    "outputId": "00475b88-9735-4486-8f14-4404772eda9e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "4.53 views, 0.17 carts, 0.07 purchases. 23,016,651 sessions.\n",
      "\n",
      "\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "f.average_number_operations(['user_session', 'event_type'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "abCr7D_2lop_"
   },
   "source": [
    "Observing the plot, it is evident that the number of *views* is far way greater than the number of *carts* and *purchases*; specifically, it is more than 26 and 64 times greater, respectively. This outcome, as expected, confirms the result discussed in the previous sub-point: customers view a lot of products, maybe many times each, before deciding to add them to the cart and making a purchase."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "dS6lW8RMloqA"
   },
   "source": [
    "## How many times, on average, a user views a product before adding it to the cart? *"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "KGu-Qm-PloqB"
   },
   "source": [
    "To solve this task, the needed attributes are `user_id`, `product_id`, `event_type` and `event_time`. Selecting the sets related to *cart* and *view* events and joining them on `user_id` and `product_id` give the desired result. It is relevant to remark that the attribute `event_time` is crucial to filter the tuples of the joint set according to the condition `event_time_view < event_time_cart` for each `(user_id, product_id)` couple. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 306
    },
    "id": "L0_Uki8lloqB",
    "outputId": "b825bb5e-9965-4440-8abd-9e0aa9488b6c"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "On average a user views a product 10.05 times before adding it to the cart.\n"
     ]
    }
   ],
   "source": [
    "f.average_views_before_cart(['user_id', 'product_id', 'event_type', 'event_time'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "rkdaSu9ZloqF"
   },
   "source": [
    "As expected by analyzing the plot above, a user views a product **10 times** before adding it to the cart. This is the reason why the *view* event overcomes the others by far."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "fIw9LNs9loqF"
   },
   "source": [
    "## What’s the probability that products added once to the cart are effectively bought? *"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "yJDDTnnXloqG"
   },
   "source": [
    "Let $C$ and $P$ be the events such that\n",
    "\n",
    "$\n",
    "C = \\{\\text{a product is added to the cart}\\}\\\\\n",
    "P = \\{\\text{a product is bought}\\}.\n",
    "$\n",
    "\n",
    "In this task, we are asked to find the probability that products added once to the cart are effectively bought, that is the conditional probability $\\mathbb{P}(P|C)$. According to the definition of conditional probability, we have\n",
    "\n",
    "$\n",
    "\\mathbb{P}(P|C)=\\frac{\\mathbb{P}(C\\cap P)}{\\mathbb{P}(C)}.\n",
    "$\n",
    "\n",
    "The probability of the event C is\n",
    "\n",
    "$\n",
    "\\mathbb{P}(C)=\\frac{\\text{# of cart tuples}}{\\text{# of tuples}}=\\frac{n_{cart}}{n_{tuples}}.\n",
    "$\n",
    "\n",
    "Similarly, the probability of the event $C\\cap P$ is\n",
    "\n",
    "$\n",
    "\\mathbb{P}(C\\cap P)=\\frac{\\text{# of cart tuples corresponding to purchase tuples}}{\\text{# of tuples}}=\\frac{n_{cart\\_purchase}}{n_{tuples}}.\n",
    "$\n",
    "\n",
    "Then, the requested probability is $\\mathbb{P}(P|C)=\\frac{n_{cart\\_purchase}}{n_{tuples}}\\times\\frac{n_{tuples}}{n_{cart}}=\\frac{n_{cart\\_purchase}}{n_{cart}}$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "E9eiH708loqG",
    "outputId": "478d1a4c-89ae-4cc8-8cb7-b777a34cb3a0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The probability that products added once to the cart are effectively bought is 0.28%.\n"
     ]
    }
   ],
   "source": [
    "f.probability_purchase(['user_session', 'product_id', 'event_type'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "LdWfktcWloqJ"
   },
   "source": [
    "As the result points out, more than **1 addition to the cart in 3** is successful. Once again, this result visually finds confirmation in the plot above."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "2LbrnXzIloqK"
   },
   "source": [
    "## What’s the average time an item stays in the cart before being removed? *"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "VUwgZacxloqL"
   },
   "source": [
    "For this task, we came up with an estimation of the average time an item stays in the cart before being removed. Indeed, despite the [description of the data](https://www.kaggle.com/mkechinov/ecommerce-behavior-data-from-multi-category-store?select=2019-Oct.csv) and according to [this punctualization](https://www.kaggle.com/mkechinov/ecommerce-behavior-data-from-multi-category-store/discussion/132470#757896), there is no *removefromcart* event. Then, observing the behaviour of some famous e-commerce stores, we assumed that the cart state lives within a session; this means that when the session expires, the cart is deleted. In other words, the cart state is session-bound. \n",
    "\n",
    "To approximate the session expiration, we pick the last operation within each session, that is the operation with the maximum timestamp within the session. Performing the timestamp difference between a cart event and the corresponding last session operation, we are able to compute the desired average time."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "yHh8HRdAloqL",
    "outputId": "3c99fbc3-6a29-42e6-a94c-e22b408e3155"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "On average, an item stays in the cart 454.93 minutes before being removed.\n"
     ]
    }
   ],
   "source": [
    "f.average_time_cart(['event_type', 'user_session', 'product_id', 'event_time'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "scm4N3lzqH4S"
   },
   "source": [
    "On average, an item stays in the cart about **8 minutes** (454.93 seconds) before being removed."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "R8IauH7XloqO"
   },
   "source": [
    "## How much time passes on average between the first view time and a purchase/addition to cart? *"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "tRiftEAAloqO"
   },
   "source": [
    "Because the question is pretty generic, we decided to analyze two different cases in order to highlight some characteristics of the customers.\n",
    "\n",
    "The first case we dealt with is considering the average time between the very first view of a user and her very first addition to the cart/purchase. In other words, we chose to measure how much time a user takes before generating a profit."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "id": "e1uHnlUvloqP",
    "outputId": "80417ecb-76be-4340-8cdb-2e73dcf8e2d7"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The average time between the first view time and a purchase/addition to cart is 257.77 hours.\n",
      "The standard deviation is 357.54 hours.\n"
     ]
    }
   ],
   "source": [
    "f.average_time_after_first_view_1(['event_type', 'event_time', 'user_id'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "PyjjMvnCloqS"
   },
   "source": [
    "To be more aware of the amount of time, we convert the result into days:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {
    "id": "gEh7yfy8loqS",
    "outputId": "0e8b9bbb-cd06-418a-ae99-405249486a5b"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The average time between the first view time and a purchase/addition to cart is 11 days.\n",
      "The standard deviation is 15 days.\n"
     ]
    }
   ],
   "source": [
    "print(f'The average time between the first view time and a purchase/addition to cart is {round(257.77 / 24)} days.')\n",
    "print(f'The standard deviation is {round(357.54 / 24)} days.')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Otm4LXBUloqV"
   },
   "source": [
    "On average, a user generates a profit **11 days** after she learnt about our business, which is pretty much! The standard deviation of **15 days** points out that there are some outliers in the delta-time distribution: probably, there are some users who view and add/buy a product immediately and others that add/buy a product after many weeks."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "2ay3boD4loqW"
   },
   "source": [
    "In the second case we analyzed, we consider the average time between the very first view of a specific product by a user and its corresponding addition to the cart/purchase. To be clear, we estimate how much time a user takes before deciding to buy a specific product she previuosly saw. We filter the tuples by the attribute `user_id` instead of `user_session` because a user may view a product for the first time during one session and add it to the cart/purchase it during another session. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "51qgoRiBloqW",
    "outputId": "f196b726-ff06-4695-bc80-9ca141350cb2",
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The average time between the first view time and a purchase/addition to cart is 2858.08 minutes.\n"
     ]
    }
   ],
   "source": [
    "f.average_time_after_first_view_2(['event_type', 'event_time', 'user_id', 'product_id'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "RyGePQ3MDu7n"
   },
   "source": [
    "Even in this case, we convert the result into a more meaningful time unit:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "pMynBduED8Tj",
    "outputId": "ff6be7f3-1f6c-46b3-e2f2-7c804d5a548a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The average time between the first view time and a purchase/addition to cart is 48 hours.\n"
     ]
    }
   ],
   "source": [
    "print(f'The average time between the first view time and a purchase/addition to cart is {round(2858.08 / 60)} hours.')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "zcL9sRxMEARH"
   },
   "source": [
    "On average, a user decides to buy a specific product after about **48** hours."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "BUON7AGvloqa"
   },
   "source": [
    "# [RQ2]  What are the categories of the most trending products overall?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "PDC70qUyloqa"
   },
   "source": [
    "We produced a plot showing the number of sold products per category: obtaining a subset of only the purchases, we grouped by category code, and counted the number of sold product plotting the biggest 10 values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 545
    },
    "id": "nIviSGOdloqb",
    "outputId": "d49d1692-590e-4d94-fae9-f58b4bc398cd"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbgAAAEICAYAAAAtAOHGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3de5xWVd338c+XkUBAOQj14HHMmyINRZw8EHpjmR20PIRpUor6xGMHT91WdFdaaj0qpaUdDE0xNTPUzLBUwlBLUWeUo4haYkWGKTgi3BnC7/5jr9HtODPMxcx1YM/3/Xpdr9l77bXX/u398uLnWntfeykiMDMzK5pe1Q7AzMysHJzgzMyskJzgzMyskJzgzMyskJzgzMyskJzgzMyskJzgzKxkkr4u6dpqx2HWESc4sxojaZmkZyX1z5X9X0lzqhiW2WbHCc6sNtUBp1U7iEqQtEW1Y7BicoIzq01TgTMlDWq9QdJYSQ9Jak5/x6byoyU1tqp7hqRb03IfSd+W9BdJKyRdJmnLtG28pL9J+mLqPT4j6XBJH5L0uKSVkv67VSh9Jd0gabWkhyXtkTvutpJukvRPSU9JOjW37euSbpR0raQXgUndddHM8pzgzGpTIzAHODNfKGkIcBtwCbANcBFwm6RtgF8Db5c0IrfLscDP0vL5wNuA0cB/ANsBZ+Xq/h+gb678cuATwF7A/sDXJO2cq38YMAMYko5xi6TeknqlWOantt4LnC7p/a32vREYBFxXwnUx6zQnOLPadRZwiqRhubJDgCci4pqIeCUirgceAz4cEWuBXwEfB0iJbiRwqyQBk4EzImJlRKwGvgUck2t7HfDNiFgH/BwYCnwvIlZHxGLgUWCPXP2miLgx1b+ILDnuC7wLGBYR50TEvyPiz2TJMn+s+yPilojYEBH/0w3XyuwNPPZtVqMiYpGkmcAUYEkq3hZ4ulXVp8l6SpD1pL4DnEPWe7slItZKejPQD2jKch0AIrvX1+L5iFiflluSzorc9v8BBuTW/5qLdYOkv6X4AthW0gu5unXAvW3ta1YuTnBmte1s4GGypAXwd2CnVnV2BG5Py7OAYZJGk/Xkzkjlz5ElqN0iYnk3xbZDy0Ialtw+xfcK8FREjGhvR7IkaFZWHqI0q2ER8SRwA9DykMZvgLdJOlbSFpKOBnYFZqb668jui00luzc2K5VvIBsmvDj15pC0Xav7YqXaS9KR6SnI04GXgbnAg8BqSV+StKWkOknvlPSuLhzLrGROcGa17xygP0BEPA8cCvwX8DzwReDQiHguV/9nwEHAjIh4JVf+JeBJYG56evF3wNu7ENevgKOBVcAngSMjYl0a5jyU7GGWp8h6j1cAA7twLLOSyROemplZEbkHZ2ZmheQEZ2ZmheQEZ2ZmheQEZ2ZmheTfwVXQ0KFDo76+vtphmJltVpqamp6LiGEbr/l6TnAVVF9fT2Nj48YrmpnZqyS1fntPp3iI0szMCskJzszMCskJzszMCskJzszMCskJroIWLm+udghmZj2GE5yZmRWSE1wXSRovaWy14zAzs9dzguuCNA/WeMAJzsysxvSoH3pL6g/8gmzm4TrgXOCCVPZBshmPj42IJyXVA1cCQ4F/AidExF8kTQf+BewJLCdLbuslfQI4JSLureQ5mZlZ23paD+4DwN8jYo+IeCdweypvjohRwPeB76ayS4GrI2J34Drgklw72wNjI+JI4DLg4ogY3VZykzRZUqOkxvVr/ZCJmVml9LQEtxB4n6QLJO0fES0Z5/rc3/3S8n5kMyMDXAOMy7UzI81avFERMS0iGiKioa6fJzQ2M6uUHjVEGRGPSxoDfAg4T9Lslk35ap1oak23B2dmZt2qR/XgJG0LrI2Ia4GpwJi06ejc3/vT8n3AMWl5ItDevbXVwFbdH62ZmXVFj+rBAaOAqZI2AOuATwM3AoMlLQBeBj6e6p4CXCXpC6SHTNpp89fAjZIOww+ZmJnVDEV0ZkSuuCQtAxoi4rlyH6vP8BHx8jNPlPswZmaFIqkpIhpK3a9HDVFW26jt/JCJmVml9LQhyjeIiPpqx2BmZt3PPTgzMyskJzgzMyskJzgzMyskJzgzMyskJzgzMyskJzgzMyskJzgzMyskJzgzMyskJzgzMyskJ7gKWri8mfopt1U7DDOzHsEJzszMCskJrh2StpV0Y7XjMDOzTdPjX7bcFklbRMTfgQnVjsXMzDZNxXtwkm6R1CRpsaTJqewDkh6WNF/S7FQ2QNJVkhZKWiDpo6n8YEn3p/ozJA1I5edLejTV/XYqO0rSotTuPamsb67dRyQdmMonSbpV0l3AbEn1khalbXWSpkp6KLX//1L5cEn3SJqXjrN/hS+nmZm1oxo9uBMjYqWkLYGHJP0KuBw4ICKekjQk1fsa0BwRowAkDZY0FPgqcFBErJH0JeDzkn4AHAGMjIiQNCi1cRbw/ohYniv7LBARMUrSSOBOSW9L28YAu6f46nMxn5RieZekPsAfJd0JHAncERHflFQH9Gt9simJTwao23pY166cmZl1WjUS3KmSjkjLO5D9439PRDwFEBEr07aDgGNadoqIVZIOBXYlSzAAbwLuB5qBfwE/kTQTmJl2+yMwXdIvgJtT2Tjg0tTmY5KeBloS3Kzc8fMOBnaX1DJkORAYATwEXCmpN3BLRMxrvWNETAOmQTajdyeuj5mZdYOKJjhJ48kS134RsVbSHGAeMLKzTZAloY+30fbewHvJ7pt9DnhPRJwsaR/gEKBJ0l4baX9NB8c9JSLuaOO4B6T2p0u6KCJ+2slzMTOzMqr0PbiBwKqU3EYC+wJ9gQMk7QyQG6KcRTacSCofDMwF3i3pP1JZf0lvS/fhBkbEb4AzgD3S9l0i4oGIOAv4J1mP8V5gYtr+NmBHYOlG4r4D+HTqqZGO2V/STsCKiLgcuIJsiNPMzGpApYcobwdOlrSELKnMJUs8k4GbJfUCngXeB5wH/CA96LEe+EZE3CxpEnB9uhcG2T251cCvJPUl6219Pm2bKmlEKpsNzAceA34kaSHwCjApIl5OQ57tuQKoBx5WVvGfwOHAeOALktYBLwHHdeHamJlZN1KEbwtVSkNDQzQ2NlY7DDOzzYqkpohoKHU//9DbzMwKyQnOzMwKyQnOzMwKyQnOzMwKyQnOzMwKyQnOzMwKyQnOzMwKyQnOzMwKyQnOzMwKyQnOzMwKyTN6V9DC5c3UT7nt1fVl5x9SxWjMzIrNPTgzMyskJzgzMyskJzhAUn2alidf1iDpknbqL5M0tDLRmZnZpvA9uHZERCPguW3MzDZT7sG1Iumtkh6R9AVJM1PZNpLulLRY0hVkE6i2zCh+m6T5khZJOrqqwZuZ2auc4HIkvR24CZgEPJTbdDbwh4jYDfglsGMq/wDw94jYIyLeSTZjees2J0tqlNS4fm1zWeM3M7PXOMG9ZhjwK2BiRMxvte0A4FqAiLgNWJXKFwLvk3SBpP0j4g0ZLCKmRURDRDTU9RtYxvDNzCzPCe41zcBfgHGd3SEiHgfGkCW68ySdVabYzMysRE5wr/k3cARwnKRjW227BzgWQNIHgcFpeVtgbURcC0wlS3ZmZlYD/BRlTkSskXQoMAs4N7fpG8D1khYD95H19ABGAVMlbQDWAZ+uZLxmZtY+RUS1Y+gxGhoaorHRvzwwMyuFpKaIaCh1Pw9RmplZITnBmZlZITnBmZlZITnBmZlZITnBmZlZITnBmZlZITnBmZlZITnBmZlZITnBmZlZITnBmZlZIfldlBW0cHkz9VNue0P5svMPqUI0ZmbF5h6cmZkV0mab4CT9dze2NUjSZ3Lr20q6sbvaNzOzyttsExzQZoJTptTzGgS8muAi4u8RMaErwZmZWXWVlAgkjZN0QloeJmnnTuxznKQFkuZLukZSvaS7UtlsSTumetMlXSLpPkl/ljQhlQ+XdI+keZIWSdpf0vnAlqnsutTmUkk/BRYBO0h6KRfDBEnT0/JbJP0yxTNf0ljgfGCX1N7U1N6iVL+vpKskLZT0iKQDU/kkSTdLul3SE5IuLOVamplZeXX6IRNJZwMNwNuBq4DewLXAuzvYZzfgq8DYiHhO0hDgauDqiLha0onAJcDhaZfhwDhgJHArcCPZTNp3RMQ3JdUB/SLiXkmfi4jR6Tj1wAjg+IiYm8raC+sS4O6IOCK1NwCYAryzVXstPgtERIySNBK4U9Lb0rbRwJ7Ay8BSSZdGxF9bXYPJwGSAuq2HtReTmZl1s1J6cEcAHwHWQDaMB2y1kX3eA8yIiOfSPiuB/YCfpe3XkCW0FrdExIaIeBR4Syp7CDhB0teBURGxup1jPd2S3DoR049SPOsjonkj9ceRJXIi4jHgaaAlwc2OiOaI+BfwKLBT650jYlpENEREQ12/gZ0Iz8zMukMpCe7fkU3/HQCS+pchnpdzywKIiHuAA4DlwHRJx7Wz75pW6/mpyvt2W4Svl493Pf7ZhZlZzSglwf1C0o+BQZI+BfwOuHwj+9wFHCVpG4A0RHkfcEzaPhG4t6MGJO0ErIiIy4ErgDFp0zpJvTvYdYWkd6QHTo7Ilc8GPp3arpM0EFhN+73Re1OcpKHJHYGlHcVsZmbV1+kEFxHfJrsndhPZfbizIuLSjeyzGPgmcLek+cBFwClkQ44LgE8Cp23k0OOB+ZIeAY4GvpfKpwELJF3Xzn5TgJlkCfWZXPlpwIGSFgJNwK4R8Tzwx/QQy9RW7fwQ6JXq3wBMioiXMTOzmqZs1NEqoc/wETH8+O++odxvMjEza5+kpohoKHW/jd4zkrSa19/Pep2I2LrUg/ZUo7YbSKOTmZlZRWw0wUXEVgCSziUb6ruG7AGQiWSP9ZuZmdWcUh4y+UhE/DAiVkfEixHxI+CwcgVmZmbWFaUkuDWSJqYnD3tJmsgbH803MzOrCaUkuGOBjwErgGeBo1KZmZlZzen0D5MjYhkekjQzs81Ep3twkrZPLyl+Nn1ukrR9OYMzMzPbVKUMUV5F9gLkbdPn16nMzMys5pSS4IZFxFUR8Ur6TAf8enwzM6tJpSS45yV9Ij1FWSfpE8Dz5QrMzMysK0pJcCeSPUX5D7IffE8ATihHUGZmZl3ld1FWUHvvogS/j9LMrD2b+i7KUp6ivFrSoNz6YElXlnpAMzOzSihliHL3iHihZSUiVgF7dn9ItUnS4ZJ2rXYcZmbWOaUkuF6SBrespMlLe9IM1ocDTnBmZpuJUhLcd4D7JZ2bZha4D7iwPGFVhqRbJDVJWixpcip7Kbd9gqTpksYCHwGmSponaRdJoyXNlbQg/QB+cHvHMTOzyitlRu+fAkeSvYtyBXBkRFzTsn0z/Qf+xIjYC2gATpW0TVuVIuI+sh+5fyEiRkfEn4CfAl+KiN2BhcDZbe0rabKkRkmN69c2l+cszMzsDUoaYoyIR4FH29k8GxjT5Ygq61RJR6TlHYARndlJ0kBgUETcnYquBma0VTcipgHTIHuKsmvhmplZZ3XnPTR1Y1tlJ2k8cBCwX0SslTQH6MvrZy/vW4XQzMysG5RyD25jNrfeyUBgVUpuI4F9U/kKSe+Q1As4Ild/NbAVQEQ0A6sk7Z+2fRK4GzMzqxndmeA2N7cDW0haApwPzE3lU4CZZA/RPJOr/3PgC5IekbQLcDzZQycLgNHAORWL3MzMNqrHDlFGxMvAB9vZfGMb9f/IG38msG/remZmVhs6neAkfQe4MiIWt1Plvd0TUnGN2m4gjX4ll5lZRZQyRLkEmCbpAUknpycJXxURK7s3NDMzs01Xyu/groiIdwPHAfXAAkk/k3RguYIzMzPbVCU9ZCKpDhiZPs8B84HPS/p5GWIzMzPbZKXcg7sYOBS4C/hWRDyYNl0gaWk5gjMzM9tUnUpwkgSsBEZHxJo2quzdrVGZmZl1UaeGKCObFfVj7SS3lh8+m5mZ1YxS7sE9LOldZYvEzMysG5XyQ+99gImSngbWkP2wO9Lb9M3MzGpKKQnu/WWLwszMrJuV8ju4p4FBwIfTZ1AqMzMzqzml/EzgNOBTwM2p6FpJ0yLi0rJEVkALlzdTP+W2drcv82u8zMy6TSlDlCcB+7Q8SSnpAuB+wAnOzMxqTilPUQpYn1tfTwVnEJB0qqQlkq7rhrbOkXRQWj5dUr+uR2hmZrWklB7cVcADkn6Z1g8HftL9IbXrM8BBEfG3jVVMP0xXRGxoa3tEnJVbPR24Fljb2UAkbRERr3S2vpmZVV4pD5lcBJxA9kaTlcAJEfHdcgWWJ+ky4K3AbyU1Szozt22RpPr0WSrpp8AiYP/U47tc0mJJd0raMu0zXdIESacC2wK/l/T7tO2lXNsTJE3P7XOZpAeACyXtIul2SU2S7k2zgpuZWY0o5SGTIcCy9Gkp6x0R67o/rNeLiJMlfQA4EPhcB1VHAMdHxFxJ9Wn94xHxKUm/AD5K1ltrafcSSZ8HDoyI5zoRyvbA2IhYL2k2cHJEPCFpH+CHwHta7yBpMjAZoG7rYZ04hJmZdYdShigfBnYAVpHdexsE/EPSCuBTEdFUhvhK9XREzM2tPxUR89JyE9k0P10xIyW3AcBYYEY2GgpAn7Z2iIhpwDSAPsNHRBePb2ZmnVRKgpsF3BgRdwBIOpisR3QVWe9ln+4Pr02v8Pqh1b655dbvynw5t7we2LIT7eeTUN9W21ra7wW8EBGjO9GemZlVQSlPUe7bktwAIuJOYL/UY2qz91Imy4AxAJLGADt3sb3VwFa59RWS3iGpF3BEWztExIvAU5KOSnFI0h5djMPMzLpRKQnuGUlfkrRT+nyRLBnUAW0+rVgmNwFDJC0mux/3eBfbmwbc3vKQCTAFmAncBzzTwX4TgZMkzQcWA4d1MQ4zM+tGymbC6URFaShwNjCObBjvj8A5QDOwY0Q8Wa4gi6KhoSEaGxurHYaZ2WZFUlNENJS6X6fvwaWnDE+R1L+NeeGc3MzMrKZ0eohS0lhJjwJL0voekn5YtsjMzMy6oJR7cBeTTZnzPEBEzAcOKEdQZmZmXVVKgiMi/tqqaH2bFc3MzKqslN/B/VXSWCAk9QZOIw1XmpmZ1ZpSenAnA58FtgOWA6PJXoBsZmZWc0rpwb09IibmCyS9m+znAmZmZjWllB5cWxOberJTMzOrSRvtwUnaj+zFwsPSm/dbbA3UlSswMzOzrujMEOWbgAGpbv6djS8CE8oRlJmZWVeV8qqunSLi6TLHU2h9ho+I4cd3PEfssvMPqVA0Zmabh7K/qgtYK2kqsBu5aWQi4g2TfJqZmVVbKQ+ZXAc8RjY9zTfIpq15qAwxmZmZdVkpCW6biPgJsC4i7o6IEwH33gBJ4yXNrHYcZmb2mlKGKNelv89IOgT4OzCk+0OqTZK2iIhXqh2HmZl1Tik9uPMkDQT+CzgTuAI4vSxRlUDSLZKaJC2WNDmVvSTp4lQ2W9KwVD5H0vckzZO0SNLeqXxvSfdLekTSfZLensonSbpV0l3AbEn9JV0p6cFU15OcmpnVqFIS3FFkT10uiogDgfcBR5QnrJKcGBF7AQ3AqZK2AfoDjRGxG3A32UStLfpFRMtrxq5MZY8B+0fEnsBZwLdy9ccAEyLiP4GvAHdFxN7AgcBUSf07Ck7SZEmNkhrXr23u8smamVnnlDJEuXtEvNCyEhErJe1ZhphKdaqklkS7AzAC2ADckMquBW7O1b8eICLukbS1pEFkv++7WtIIstnKe+fqz4qIlWn5YOAjks5M632BHTsKLiKmAdMg+5nAJpyfmZltglISXC9JgyNiFYCkISXu3+0kjQcOAvaLiLWS5pD7CUNOtLPcsn4u8PuIOEJSPTAntz0/e7mAj0bE0lZxvGUTwjczszIqZYjyO8D9ks6VdC5wH3BhecLqtIHAqpTcRgL7pvJevPaWlWOBP+T2ORpA0jigOSKaUzvL0/ZJHRzvDuAUSUpt1EIP1szM2tDpBBcRPwWOBFakz5ERcU25Auuk24EtJC0BzgfmpvI1wN6SFpH9lOGc3D7/kvQIcBlwUiq7EPj/qbyjXum5ZMOXCyQtTutmZlaDOv2qrs2JpJciYkAb5XOAMyOisfJRQUNDQzQ2VuXQZmabrU19VVcpQ5RmZmabjao+JFIubfXeUvn4CodiZmZV4h6cmZkVkhOcmZkVkhOcmZkVkhOcmZkVkhOcmZkVkhOcmZkVkhOcmZkVkhOcmZkVUiF/6F2rFi5vpn7KbWVpe9n5h5SlXTOzzZV7cGZmVkhOcICk0ZI+VO04zMys+zjBZUYDJSU4SR7eNTOrYTWR4CQdJ2mBpPmSrpFUL+muVDZb0o6p3nRJP5I0V9KfJY2XdKWkJZKm59p7SdLFkhan/Yel8jmSGtLyUEnLJL2JbL64oyXNk3S0pP6p3QclPSLpsLTPJEm3SroLmC1puKR70n6LJO1f6WtnZmZtq3qCk7Qb8FXgPRGxB3AacClwdUTsDlwHXJLbZTCwH3AGcCtwMbAbMErS6FSnP9AYEbsBdwNnt3f8iPg3cBZwQ0SMjogbgK8Ad0XE3sCBwFRJ/dMuY4AJEfGfZLOF3xERo4E9gHltnN9kSY2SGtevbS718piZ2SaqeoIjm3F7RkQ8BxARK8kS2M/S9muAcbn6v45sltaFwIqIWBgRG4DFQH2qswG4IS1f22r/zjgYmCJpHjAH6AvsmLbNSjECPAScIOnrwKiIWN26oYiYFhENEdFQ129giWGYmdmmqoUEV6qX098NueWW9fbui7VMW/4Kr51z3w6OIeCjqUc3OiJ2jIgladuaVxuNuAc4AFgOTJd0XOdPw8zMyqkWEtxdwFGStgGQNAS4DzgmbZ8I3Ftim72ACWn5WOAPaXkZsFdanpCrvxrYKrd+B3CKJKWY9mzrIJJ2IutFXg5cQTZ8aWZmNaDqCS4iFgPfBO6WNB+4CDiFbOhvAfBJsvtypVgD7C1pEdkQ6Dmp/NvApyU9AgzN1f89sGvLQybAuUBvYIGkxWm9LeOB+am9o4HvlRinmZmVibLbWcUi6aWIGFDtOFrrM3xEDD/+u2Vp228yMbOiktQUEQ2l7uffclXQqO0G0uhEZGZWEVUfoiyHWuy9mZlZZRUywZmZmTnBmZlZITnBmZlZITnBmZlZITnBmZlZITnBmZlZITnBmZlZITnBmZlZITnBmZlZIflVXRW0cHkz9VNuq3YYVeP3ZZpZJbkHZ2ZmhVSIBCdpkqTvp+WTPfGomZkVbogyIi6rdgxmZlZ9Ve/BSbpFUpOkxZImp7KXJF2cymZLGpbK50j6XpqYdJGkvdto7+uSzkzLn5L0kKT5km6S1C+VT5d0iaT7JP1Z0oTc/l+StDDtc34q20XS7SnOeyWNTOVHpTjmS7qn/FfLzMw6q+oJDjgxIvYCGoBTJW0D9AcaI2I34G7g7Fz9fhExGvgMcOVG2r45It4VEXsAS4CTctuGA+OAQ4GWRPZB4DBgn7TPhanuNOCUFOeZwA9T+VnA+1Pdj7QVgKTJkholNa5f27yxa2FmZt2kFoYoT5V0RFreARgBbABuSGXXAjfn6l8PEBH3SNpa0qAO2n6npPOAQcAA4I7ctlsiYgPwqKS3pLKDgKsiYm06xkpJA4CxwAxJLfv2SX//CEyX9ItWMb4qIqaRJUj6DB9RvOnTzcxqVFUTnKTxZEllv4hYK2kO0LeNqtHOclvredOBwyNivqRJwPjctpfzoXTQRi/ghdRrfP2BI06WtA9wCNAkaa+IeL6DtszMrEKqPUQ5EFiVkttIYN9U3gtouS92LPCH3D5HA0gaBzRHREfjflsBz0jqDUzsRDyzgBNy9+qGRMSLwFOSjkplkrRHWt4lIh6IiLOAf5L1QM3MrAZUe4jyduBkSUuApcDcVL4G2FvSV4FnSUkt+ZekR4DewIkbaf9rwANkyecBsoTXroi4XdJooFHSv4HfAP9Nlhx/lOLpDfwcmA9MlTSCrAc4O5WZmVkNUETt3RaS9FJEDGijfA5wZkQ0Vj6qruszfEQMP/671Q6javwmEzPbFJKaIqKh1P2q3YPrUUZtN5BG/yNvZlYRNZng2uq9pfLxFQ7FzMw2U9V+yMTMzKwsnODMzKyQnODMzKyQnODMzKyQnODMzKyQnODMzKyQnODMzKyQnODMzKyQavKH3kW1cHkz9VNuq3YYZmYVVa3X9LkHZ2ZmheQEZ2ZmhVRzCU7SMklDN2G/8ZLGdmMcDZIu6a72zMyssop0D2488BJwX+sNkraIiFdKaSxNybNZTstjZmZV7sFJ+oSkByXNk/RjSXWd2S7pA5IeljRf0mxJ9cDJwBmp7v6Spku6TNIDwIWSRkuaK2mBpF9KGpzamiPpgnScxyXtn8rHS5qZlgdIukrSwrT/RyXVpWMsSuVnVPDSmZnZRlStByfpHWQzdb87ItZJ+iHZzNkdbpf0W+By4ICIeErSkIhYKeky4KWI+Hba/yRge2BsRKyXtAA4JSLulnQOcDZwejrcFhGxt6QPpfKDWoX7NaA5IkaltgcDo4HtIuKdqWxQO+c5GZgMULf1sK5cMjMzK0E1hyjfC+wFPCQJYEvg2U5s3xe4JyKeAoiIlR0cY0ZKbgOBQRFxdyq/GpiRq3dz+tsE1LfRzkHAMS0rEbFK0p+Bt0q6FLgNuLOtACJiGjANshm9O4jVzMy6UTUTnICrI+LLryuUJm1k+4dLOMaaTtZ7Of1dTyevSUpyewDvJxse/RhwYgmxmZlZGVXzHtxsYIKkNwNIGiJpp05snwscIGnnlvJUfzWwVVsHiohmYFXL/TXgk8DdbdVtxyzgsy0rkganJz17RcRNwFeBMSW0Z2ZmZVa1HlxEPCrpq8CdknoB68glkfa2R8TcdF/r5lT+LPA+4NfAjZIOA05p45DHA5dJ6gf8GTihhHDPA34gaRFZL+8bwJ+Aq1IMAF9ub2czM6s8Rfi2UKU0NDREY6N/eWBmVgpJTRHRUOp+NfdDbzMzs+7gBGdmZoXkBGdmZoXkBGdmZoXkBGdmZoXkpygrSNJqYGm146iyocBz1Q6iinr6+YOvAfgalHr+O0VEye86LNJsApuDpZvyqGuRSGrsydegp58/+BqAr0Glzt9DlGZmVkhOcGZmVkhOcJU1rdoB1ICefg16+vmDryAs0doAAAR6SURBVAH4GlTk/P2QiZmZFZJ7cGZmVkhOcGZmVkhOcBUi6QOSlkp6UtKUasdTKklXSno2TRnUUjZE0ixJT6S/g1O5JF2SznWBpDG5fY5P9Z+QdHyufC9JC9M+lyhN497eMSpN0g6Sfi/pUUmLJZ3WUXwFvQZ9JT0oaX66Bt9I5TtLeiDFfYOkN6XyPmn9ybS9PtfWl1P5Uknvz5W3+T1p7xjVIKlO0iOSZnYUW4HPf1n673SepMZUVpvfg4jwp8wfoI5s/ri3Am8C5gO7VjuuEs/hALJJXRflyi4EpqTlKcAFaflDwG/JZmXfF3gglQ8hm4tvCDA4LQ9O2x5MdZX2/WBHx6jC+Q8HxqTlrYDHgV172DUQMCAt9wYeSPH+AjgmlV8GfDotfwa4LC0fA9yQlndN34E+wM7pu1HX0fekvWNU6Tp8HvgZMLOj2Ap8/suAoa3KavJ7UJUL1NM+wH7AHbn1LwNfrnZcm3Ae9bw+wS0Fhqfl4WQ/ZAf4MfDx1vWAjwM/zpX/OJUNBx7Llb9ar71jVPsD/Ipsot0eeQ2AfsDDwD5kb6TYIpW/+t86cAewX1reItVT6//+W+q19z1J+7R5jCqc9/bAbOA9wMyOYivi+afjL+ONCa4mvwceoqyM7YC/5tb/lso2d2+JiGfS8j+At6Tl9s63o/K/tVHe0TGqJg017UnWg+lR1yANz80DngVmkfU4XoiIV1KVfNyvnmva3gxsQ+nXZpsOjlFp3wW+CGxI6x3FVsTzBwjgTklNkianspr8HvhVXdYtIiIklfU3J5U4xsZIGgDcBJweES+m2wNAz7gGEbEeGC1pEPBLYGS1Yqk0SYcCz0ZEk6Tx1Y6nisZFxHJJbwZmSXosv7GWvgfuwVXGcmCH3Pr2qWxzt0LScID099lU3t75dlS+fRvlHR2j4iT1Jktu10XEzRuJr5DXoEVEvAD8nmy4bJCklv9Zzsf96rmm7QOB5yn92jzfwTEq6d3ARyQtA35ONkz5vQ5iK9r5AxARy9PfZ8n+J2dvavR74ARXGQ8BI9KTUG8iu+F8a5Vj6g63Ai1PPx1Pdl+qpfy49ATVvkBzGlq4AzhY0uD0BNTBZPcSngFelLRvemLquFZttXWMikpx/QRYEhEX5Tb1pGswLPXckLQl2T3IJWSJbkIb8eXjngDcFdkNlFuBY9JThjsDI8geLGjze5L2ae8YFRMRX46I7SOiPsV2V0RM7CC2Qp0/gKT+krZqWSb773cRtfo9qNaNyp72IXua6HGyexZfqXY8mxD/9cAzwDqycfGTyO4NzAaeAH4HDEl1BfwgnetCoCHXzonAk+lzQq68IX1R/gR8n9festPmMapw/uPI7j0sAOalz4d62DXYHXgkXYNFwFmp/K1k/0A/CcwA+qTyvmn9ybT9rbm2vpLOcynpKbmOviftHaOK34fxvPYUZY85/xTH/PRZ3BJjrX4P/KouMzMrJA9RmplZITnBmZlZITnBmZlZITnBmZlZITnBmZlZITnBmZlZITnBmZlZIf0vN556Aiiw5hEAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "f.num_sold_per_cat()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "VUQAPfptloqe"
   },
   "source": [
    "We then plotted the 10 most visited subcategories, obtaining a subset of only the views, grouping by category code and counting the views of each category"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 571
    },
    "id": "8jvDV6Igloqe",
    "outputId": "1212c248-4029-4567-dcde-aaa3512f187d"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "f.most_v_cat()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "xHSyUYvrloqh"
   },
   "source": [
    "And eventually we produced a list showing the most sold products per category: obtaining a subset of only the purchases, we counted, for each category code, the number of sold products"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "HHoay2UYloqh",
    "outputId": "2de7789e-f8fc-4809-8e38-574095bdd681"
   },
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/usr/local/lib/python3.6/dist-packages/pandas/core/generic.py:5168: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  self[name] = value\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "accessories\n",
      "product_id\n",
      "18300155    105\n",
      "18300021     87\n",
      "28401112     56\n",
      "18300460     51\n",
      "28400759     50\n",
      "18300141     42\n",
      "18300214     39\n",
      "52900016     38\n",
      "18300370     33\n",
      "28400912     31\n",
      "Name: event_type, dtype: int64\n",
      "apparel\n",
      "product_id\n",
      "28720716    101\n",
      "54900013    100\n",
      "54900004     97\n",
      "28718083     93\n",
      "28716978     91\n",
      "54900011     89\n",
      "28716666     85\n",
      "28719606     83\n",
      "28719635     74\n",
      "54900012     71\n",
      "Name: event_type, dtype: int64\n",
      "appliances\n",
      "product_id\n",
      "3700926    3439\n",
      "3600661    3224\n",
      "3600666    2008\n",
      "3601405    1908\n",
      "3600163    1896\n",
      "3700766    1384\n",
      "2900958    1235\n",
      "3601485    1189\n",
      "3701134    1167\n",
      "2702277    1100\n",
      "Name: event_type, dtype: int64\n",
      "auto\n",
      "product_id\n",
      "6000094    1885\n",
      "4700478     894\n",
      "6000227     849\n",
      "5701128     628\n",
      "5701166     578\n",
      "6000004     480\n",
      "6000229     467\n",
      "4700630     452\n",
      "4700589     397\n",
      "5701002     341\n",
      "Name: event_type, dtype: int64\n",
      "computers\n",
      "product_id\n",
      "1307310    2332\n",
      "1307073    1604\n",
      "1307188    1207\n",
      "1307067    1189\n",
      "1306650     924\n",
      "1307545     908\n",
      "1307366     763\n",
      "1307187     744\n",
      "1307004     696\n",
      "1307076     683\n",
      "Name: event_type, dtype: int64\n",
      "construction\n",
      "product_id\n",
      "19200005    741\n",
      "12300396    554\n",
      "4000169     537\n",
      "30000048    362\n",
      "4000170     356\n",
      "12300066    267\n",
      "12400121    256\n",
      "12300394    229\n",
      "19200010    190\n",
      "30000055    181\n",
      "Name: event_type, dtype: int64\n",
      "country_yard\n",
      "product_id\n",
      "27600080    24\n",
      "27600022    17\n",
      "30500002    10\n",
      "30500003     9\n",
      "30500001     7\n",
      "30500022     6\n",
      "27600165     5\n",
      "30500023     4\n",
      "33700006     4\n",
      "27600024     3\n",
      "Name: event_type, dtype: int64\n",
      "electronics\n",
      "product_id\n",
      "1004856    61265\n",
      "1004767    44419\n",
      "1005115    34787\n",
      "4804056    30181\n",
      "1004833    26183\n",
      "1002544    22227\n",
      "1004870    21288\n",
      "1004249    17971\n",
      "1005105    15776\n",
      "1004836    15549\n",
      "Name: event_type, dtype: int64\n",
      "furniture\n",
      "product_id\n",
      "14701435    453\n",
      "17200570    343\n",
      "17200651    256\n",
      "7100133     226\n",
      "14701260    192\n",
      "7101172     162\n",
      "14701406    159\n",
      "7900440     158\n",
      "13200019    147\n",
      "13201002    147\n",
      "Name: event_type, dtype: int64\n",
      "kids\n",
      "product_id\n",
      "7004492     435\n",
      "7002254     267\n",
      "7004508     232\n",
      "8902633     168\n",
      "7004509     164\n",
      "12100045    164\n",
      "7005751     156\n",
      "7006453     144\n",
      "9002789     141\n",
      "8900305     136\n",
      "Name: event_type, dtype: int64\n",
      "medicine\n",
      "product_id\n",
      "25800007    196\n",
      "25800005    193\n",
      "25800003     82\n",
      "25800001     39\n",
      "25800016     34\n",
      "25800002     31\n",
      "25800006     18\n",
      "25800020     15\n",
      "25800004     13\n",
      "25800014      6\n",
      "Name: event_type, dtype: int64\n",
      "sport\n",
      "product_id\n",
      "26900021    145\n",
      "12202301     99\n",
      "12201562     93\n",
      "26900065     71\n",
      "26900009     60\n",
      "26900059     57\n",
      "26900073     57\n",
      "12201563     51\n",
      "12200545     49\n",
      "12201949     48\n",
      "Name: event_type, dtype: int64\n",
      "stationery\n",
      "product_id\n",
      "53400013    26\n",
      "12900110    13\n",
      "12901271    12\n",
      "12901426    11\n",
      "12900125    10\n",
      "12900132    10\n",
      "12900142    10\n",
      "12900360    10\n",
      "12900363    10\n",
      "53400004     9\n",
      "Name: event_type, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "f.most_sold_per_cat()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "pf2yOb7Qtx7E"
   },
   "source": [
    "# [RQ3] For each category, what’s the brand whose prices are higher on average?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "NdtpSzOOuGIc"
   },
   "source": [
    "As a result of the analysis, what appears to be a constant trend throughout the answers, is that the most expensive brands, and associated products as well, are by far the ones of electronics devices and electronic domestic appliances; the only exception consists of sport bycicles outpricing the others. To have a general overview of the average price by brand for each category we implemented this function, it is easy to observe the brands whose prices are higher on average."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "uNthyXXduPtl",
    "outputId": "8661ec81-9af0-450e-c33a-b75864d0467a"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "category_code  brand        \n",
      "sport          pinarello        2533.238525\n",
      "               nordictrack      2522.590088\n",
      "               kessler          2372.000000\n",
      "appliances     climadiff        2261.951172\n",
      "computers      dreammachines    2221.466309\n",
      "construction   helix            2199.047852\n",
      "electronics    rado             2062.926025\n",
      "furniture      senspa           1976.539551\n",
      "               trevi            1959.888672\n",
      "electronics    louiserard       1931.751465\n",
      "Name: price, dtype: float32 \n",
      "\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "f.avg_price_by_brand()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "_aQsr0DoIIDf"
   },
   "source": [
    "## Write a function that asks the user a category in input and returns a plot indicating the average price of the products sold by the brand"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Ivzl5j22AKLG"
   },
   "source": [
    "The function we implemented is quite similar to the one called above, except for the fact that the user is asked to indicate a precise category.\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "YRdWbYmhIeDm"
   },
   "source": [
    "```python\n",
    "def avg_price_by_brand_cat():\n",
    "    global DATA_PATH\n",
    "    cols = ['category_code', 'brand', 'price']\n",
    "    ds10 = pd.read_csv(DATA_PATH + \"2019-Oct.csv\", dtype={'price':np.float32}, usecols=cols)\n",
    "    ds11 = pd.read_csv(DATA_PATH + \"2019-Nov.csv\", dtype={'price':np.float32}, usecols=cols)\n",
    "    dstot = pd.concat([ds10, ds11]).dropna()\n",
    "\n",
    "    extract_cat = lambda x: x.split('.')[0]\n",
    "    dstot.category_code = dstot.category_code.map(extract_cat)\n",
    "\n",
    "    print(\"Enter the category_code whose average price by brand you want to know\")\n",
    "    cat = input()\n",
    "    dstot_grouped1 = dstot[dstot[\"category_code\"]==cat].groupby(\"brand\")[\"price\"].mean()\n",
    "    dstot_grouped1.plot.bar(figsize=(50, 20))\n",
    "    plt.show()\n",
    "    \n",
    "    return\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "_mtxUCczATKz"
   },
   "source": [
    "If we consider for example the category_code \"electronics\", the brand Rado turns out to have higher prices compared to the others."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 715
    },
    "id": "b4xSljJsIdIH",
    "outputId": "c18cb9e6-7f26-43d6-fb85-723de38995c5"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Enter the category_code whose average price by brand you want to know\n",
      "electronics\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "f.avg_price_by_brand_cat()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "-oXhCaR9I41Z"
   },
   "source": [
    "## Find, for each category, the brand with the highest average price. Return all the results in ascending order by price"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "GNdEiumyAbV2"
   },
   "source": [
    "Apart from some modifications, the code is resemblant to the ones above. The considerations made at the beginning of Question 3 refer particularly to this sub-question."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "I-Mhu8QlI-pL",
    "outputId": "7379b757-0894-441d-c87f-62f2e0bb21d2"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "     category_code          brand        price\n",
      "2327      medicine          omron    97.573456\n",
      "2480    stationery             hp   154.561722\n",
      "110    accessories        weekend   550.849976\n",
      "1260  country_yard         viking   712.080383\n",
      "788           auto         remeza   863.195679\n",
      "266        apparel    sergiorossi  1030.058350\n",
      "505     appliances           jura  1686.643555\n",
      "1569   electronics           rado  1758.832275\n",
      "2063          kids         hartan  1801.819946\n",
      "1896     furniture         senspa  1930.541992\n",
      "1096  construction          helix  2033.526245\n",
      "867      computers  dreammachines  2187.494873\n",
      "2422         sport    nordictrack  2522.590088\n"
     ]
    }
   ],
   "source": [
    "f.highest_avg_price_brand()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Xhoe0wYmloqk"
   },
   "source": [
    "# [RQ4] How much does each brand earn per month? *"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Nah5X-zXloql"
   },
   "source": [
    "As we were asked, we wrote the following function to get the profit per month given the brand name:"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "_jsUiedHloql"
   },
   "source": [
    "```python\n",
    "def get_profit_per_month(brand_name, dataset_list:list=None):\n",
    "    global month_files\n",
    "    if dataset_list == None:\n",
    "        dataset_list = get_purchases(['brand', 'product_id', 'price', 'event_type'])\n",
    "    profit_per_month = {}\n",
    "    for i in range(len(month_files)):\n",
    "        profit = 0\n",
    "        month_name = month_files[i][5:8]\n",
    "        brand_items = dataset_list[i][dataset_list[i].brand == brand_name]\n",
    "        if not brand_items.empty:\n",
    "            profit = brand_items[(brand_items.price.notna()) & \n",
    "                                 (brand_items.price.notnull())].price.agg('sum')\n",
    "        profit_per_month[month_name] = profit\n",
    "    return profit_per_month\n",
    "```"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "IrYbuKaSloqm"
   },
   "source": [
    "The profit of a brand is equal to the sum of the purchases of the brand products. So, we get the *purchase* tuples and we filter on the brand name. This way, the function returns a dictionary containing (month->profit) couples. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "YahRSycvloqm"
   },
   "source": [
    "For example, the profit of the brand Samsung is:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "jwxF2wexloqn",
    "outputId": "8cc2d616-4eaa-4aa4-b0f4-47b577d495cb"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Samsung earned 46,407,524.00$ in October and 54,869,880.00$ in November.\n"
     ]
    }
   ],
   "source": [
    "profit = f.get_profit_per_month('samsung')\n",
    "print(f'Samsung earned {profit[\"Oct\"]:,.2f}$ in October and {profit[\"Nov\"]:,.2f}$ in November.')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "g_hUzFV0loqp"
   },
   "source": [
    "## Is the average price of products of different brands significantly different? *"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "TlE6TwcFloqq"
   },
   "source": [
    "To answer to this question, we first compute the mean of the price of the distinct products for each brand; then, we calculate the standard deviation of the means. Indeed, the standard deviation allows us to understand how much the average price of each brand differ from the mean."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "m_JpRRUNloqq",
    "outputId": "222da51e-cf79-4130-bef3-e4caaa07208e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The mean of the average price of the products for each brand is 171.71$,\n",
      "with a standard deviation of 267.11$.\n"
     ]
    }
   ],
   "source": [
    "brand_means = f.price_std_dev(['brand', 'product_id', 'price'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "6-e_BuOPloqt"
   },
   "source": [
    "Comparing the mean we obtained with the related standard deviation, we can deduce that the variation inside the average price distribution is pretty relevant. To prove this, we can take the maximum and the minimum average price:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "c9xZuYiLloqu",
    "outputId": "101e72bd-c20f-4569-faa6-fb4a52f77885"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The minimum average price is:\t0.97$.\n",
      "The maximum average price is:\t2,522.59$.\n"
     ]
    }
   ],
   "source": [
    "print(f'The minimum average price is:\\t{brand_means.min():,.2f}$.\\nThe maximum average price is:\\t{brand_means.max():,.2f}$.')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "TsqxvMV8loqx"
   },
   "source": [
    "As shown, there is a huge difference between the maximum and the minimum average price."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "GfWOIzlNloqx"
   },
   "source": [
    "## Find the top 3 brands that have suffered the biggest losses in earnings between one month and the next *"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "h8cpvGi-loqy"
   },
   "source": [
    "We assumed as \"biggest losses\" the percentage losses between one month and the next. We compute the loss as:\n",
    "\n",
    "$\n",
    "l_{\\%} = \\frac{profit_{nov}-profit_{oct}}{profit_{oct}+1}\\times 100\n",
    "$\n",
    "\n",
    "We add 1 to $profit_{oct}$ to avoid zero divisions. This implies that, when $profit_{nov}$ is 0, the loss is *almost* 100%, but not exactly 0. We assume that this approximation is negligible."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "EQdcNvaYloqy",
    "outputId": "fda6a28f-225b-494e-c487-053de1cb2ae8"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The brand \"eltreco\" lost 99.99% between Oct and Nov.\n",
      "The brand \"kingsong\" lost 99.98% between Oct and Nov.\n",
      "The brand \"erdenet\" lost 99.98% between Oct and Nov.\n"
     ]
    }
   ],
   "source": [
    "f.top_n_two_months_losses(['brand', 'product_id', 'price', 'event_type'], 'Oct', 'Nov')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "famXbQBEloq1"
   },
   "source": [
    "As expected, the top 3 brands that suffered the biggest losses between October and November are minor brands, with small business. Their losses suggest that these 3 brands didn't earn anything (or almost anything) during November. Nevertheless, they may not be those which had the greatest *absolute losses*. To verify this out, we slightly modified the previous function.  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "12H4Zepbloq1",
    "outputId": "8f5538ce-a461-4264-e568-b2d415800af0"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The brand \"acer\" lost 229,413.00$ between Oct and Nov.\n",
      "The brand \"peda\" lost 171,063.44$ between Oct and Nov.\n",
      "The brand \"omabelle\" lost 145,833.72$ between Oct and Nov.\n"
     ]
    }
   ],
   "source": [
    "f.top_n_two_months_losses_abs(['brand', 'product_id', 'price', 'event_type'], 'Oct', 'Nov')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "su5r3Jcfloq5"
   },
   "source": [
    "Indeed, the list of the top 3 brand is now completely different."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Mu3uQMzSloq5"
   },
   "source": [
    "---\n",
    "# [RQ5]  In what part of the day is your store most visited?"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "nfk32XFiloq6"
   },
   "source": [
    "We created a heatmap plot that, for each day of the week, shows the hourly average of visitors the store has"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 322
    },
    "id": "kZ9pLlYJMvfQ",
    "outputId": "5914bfd8-9934-427c-afb8-14e39918c830"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 4 Axes>"
      ]
     },
     "metadata": {
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "f.havg_visit()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "ZlVXwQYiJ8Ar"
   },
   "source": [
    "# [RQ6] What's the conversion rate of your online store?\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "WnW_HeQDKJDN"
   },
   "source": [
    "As online store we considered as usual October and November datasets; for each month we took into account the number of purchases and the number of products only viewed, since the conversion rate is defined as the ratio between these two numbers."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "8u2BpNMHKNrB",
    "outputId": "c218adc0-478d-4e11-f4e2-bb3278d9a9b3"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Conversion rate of online store (Oct. and Nov.) : 0.01590817944828352 \n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "f.conv_rate_store()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Ym1GGQZKKbz_"
   },
   "source": [
    "## Plot the number of purchases of each category"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "bkVEJxEIAsY7"
   },
   "source": [
    "The plot shows once again the trend seen in Question 3 where the most preferred products are found in categories such as electronic devices and appliances."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/",
     "height": 678
    },
    "id": "ZjPh-nojMrUn",
    "outputId": "a20f21ba-fd6b-41a4-e06c-d1770ef66906"
   },
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1296x720 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light",
      "tags": []
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "f.purchases_cat()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "2UdpwHGCA8vA"
   },
   "source": [
    "## Show the conversion rate of each category in decreasing order"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "id": "-X4_Y4zrA7ld",
    "outputId": "da9c1be8-bf53-4dc1-c1ae-00f2d9b0e0ad"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "category_code\n",
      "electronics     0.024757\n",
      "medicine        0.018827\n",
      "stationery      0.016819\n",
      "appliances      0.013555\n",
      "computers       0.009897\n",
      "auto            0.009890\n",
      "construction    0.009376\n",
      "kids            0.009015\n",
      "sport           0.006565\n",
      "accessories     0.006068\n",
      "furniture       0.006022\n",
      "apparel         0.004962\n",
      "country_yard    0.004223\n",
      "dtype: float64\n"
     ]
    }
   ],
   "source": [
    "f.cv_rate_cat()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "KErmxNNKloq6"
   },
   "source": [
    "# [RQ7] Prove that the Pareto principle applies to your store *"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "VI64fn0hloq7"
   },
   "source": [
    "According to the Pareto principle, around 80% of the business of the e-commerce platform comes from about 20% of the customers. The keypoint in this task is state a definition of the \"platform business\". We gave two definitions of it:\n",
    "\n",
    "1. the total profit coming from the purchases.\n",
    "2. the number of operations performed within the platform.\n",
    "\n",
    "While the first case is self-explanatory, the second case needs a little explanation. We considered that the business of an e-commerce platform, that is its profit, may not come only from the purchases, but also from the activities within it; actually, even *view* events may generate some profit thanks to the ads that may be present inside the e-commerce web pages or thanks to the visibility of the products, for which each brand may pay a fee."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "45kZ6DTqloq8"
   },
   "source": [
    "According to the previous introduction, in the first case we consider the 20% of the users that have spent the most in the e-commerce platform and we compute the rate between the profit coming from them and the total profit."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "KGQ4pRRcloq8",
    "outputId": "e86a08f8-e508-4900-e1b5-7b13783a6072"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The 20% of the users is responsible of the 72.0% of the business.\n"
     ]
    }
   ],
   "source": [
    "f.prove_pareto(['event_type', 'user_id', 'price'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "r5tdRsvIloq_"
   },
   "source": [
    "In the second case, we just count how many operations each user performed, sorting them in descent order. Then, we compute the rate between the amount of operations performed by the 20% of the users and the total number of operations performed within the system. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "id": "r9u7Vjd1loq_",
    "outputId": "d3284d04-7e8f-4d2c-9fe1-7da65477082e"
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The 20% of the users is responsible of the 77.0% of the business.\n"
     ]
    }
   ],
   "source": [
    "f.prove_pareto_ops(['user_id'])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "L3Cvl4-4lorC"
   },
   "source": [
    "In both cases, the Pareto principle applies to our store."
   ]
  }
 ],
 "metadata": {
  "colab": {
   "collapsed_sections": [
    "g_hUzFV0loqp",
    "GfWOIzlNloqx",
    "KErmxNNKloq6"
   ],
   "name": "main.ipynb",
   "provenance": [],
   "toc_visible": true
  },
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}