EDA.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Unit Testing for Data Science (Berlin WiMLDS Workshop)\n",
    "## Exploratory Data Analysis\n",
    "\n",
    "This notebook contains exploratory data analysis of a dataset based on the [Pockets dataset from The Pudding](https://github.com/the-pudding/pockets). In order to try out different imputation methods we removed 10% of the price values.\n",
    "\n",
    "You can use this notebook to get an overview of the dataset or write functions for the exercises, before transfering them to the appropriate Python scripts."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import warnings\n",
    "\n",
    "warnings.filterwarnings(\"ignore\")\n",
    "\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from pandas_profiling import ProfileReport\n",
    "\n",
    "from matplotlib.ticker import PercentFormatter\n",
    "plt.rcParams.update({ \"figure.figsize\" : (8, 5),\"axes.facecolor\" : \"white\", \"axes.edgecolor\":  \"black\"})\n",
    "plt.rcParams[\"figure.facecolor\"]= \"w\"\n",
    "pd.plotting.register_matplotlib_converters()\n",
    "pd.set_option('display.float_format', lambda x: '%.3f' % x)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Index(['brand', 'style', 'menWomen', 'name', 'fabric', 'price',\n",
      "       'maxHeightFront', 'minHeightFront', 'rivetHeightFront', 'maxWidthFront',\n",
      "       'minWidthFront', 'maxHeightBack', 'minHeightBack', 'maxWidthBack',\n",
      "       'minWidthBack', 'cutout', 'waistSize', 'updatedStyle', 'group',\n",
      "       'priceGroup', 'pocketArea', 'rectanglePhone', 'rectanglePen',\n",
      "       'rectangleWallet', 'rectangleHand', 'rectangeWallet'],\n",
      "      dtype='object')\n"
     ]
    }
   ],
   "source": [
    "# read pockets dataset with 10% missing price values\n",
    "df = pd.read_csv(\n",
    "    \"data/measurementRectanglesMissing.csv\"\n",
    ")  # to not index on user id..as we got duplicates that we need\n",
    "print(df.columns)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Imputing missing values"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of missing values:  8\n"
     ]
    }
   ],
   "source": [
    "# how many missing values in 'price' column?\n",
    "print(\"Number of missing values: \", df.price.isnull().sum())\n",
    "df_impute = df.copy()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "# fucntion to impute NaNs with mean value\n",
    "def impute(series: pd.Series) -> pd.Series:\n",
    "    mean = series.mean()\n",
    "    return series.fillna(mean)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_impute[\"price\"] = impute(df_impute[\"price\"])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_impute.price.isnull().sum() "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### TODO: Exercise 1\n",
    "\n",
    "Add functions to impute missing values with min/max values, then add unit tests.\n",
    "\n",
    "Don't forget to add your functions to `imputation.py` and your unit tests to `test_imputation.py`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "# write your functions/unit tests here\n",
    "# don't forget to transfer to appropriate scripts after"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Explore the dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "#transforming categorical columns to categorical type: brand style menWomen\n",
    "df[\"brand\"] = df[\"brand\"].astype(\"category\")\n",
    "df[\"style\"] = df[\"style\"].astype(\"category\")\n",
    "df[\"menWomen\"] = df[\"menWomen\"].astype(\"category\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Are womens jeans more expensive?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    72.000\n",
       "mean     80.989\n",
       "std      45.247\n",
       "min       9.990\n",
       "25%      49.980\n",
       "50%      73.975\n",
       "75%      95.713\n",
       "max     249.000\n",
       "Name: price, dtype: float64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.price.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "# bin prices by min/max value, upper/lower percentiles\n",
    "bins = [9, 50, 74, 96, 250]\n",
    "df['price_binned'] = pd.cut(df['price'], bins)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "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>price_binned</th>\n",
       "      <th>(9, 50]</th>\n",
       "      <th>(50, 74]</th>\n",
       "      <th>(74, 96]</th>\n",
       "      <th>(96, 250]</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>menWomen</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>men</th>\n",
       "      <td>11</td>\n",
       "      <td>8</td>\n",
       "      <td>7</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>women</th>\n",
       "      <td>11</td>\n",
       "      <td>6</td>\n",
       "      <td>11</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "price_binned  (9, 50]  (50, 74]  (74, 96]  (96, 250]\n",
       "menWomen                                            \n",
       "men                11         8         7          9\n",
       "women              11         6        11          9"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# women jeans are not really more expensive\n",
    "pd.crosstab(df[\"menWomen\"], df.price_binned)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Are women pockets smaller?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count      80.000\n",
       "mean     8424.815\n",
       "std      2699.333\n",
       "min      4454.429\n",
       "25%      5905.184\n",
       "50%      8618.737\n",
       "75%     10724.592\n",
       "max     13102.032\n",
       "Name: pocketArea, dtype: float64"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.pocketArea.describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [],
   "source": [
    "# bin pocketArea by min/max value, upper/lower percentiles\n",
    "bins = [4454, 5906, 8619, 10725, 13103]\n",
    "df['pocket_binned'] = pd.cut(df['pocketArea'], bins)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "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>pocket_binned</th>\n",
       "      <th>(4454, 5906]</th>\n",
       "      <th>(5906, 8619]</th>\n",
       "      <th>(8619, 10725]</th>\n",
       "      <th>(10725, 13103]</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>menWomen</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>men</th>\n",
       "      <td>0.000</td>\n",
       "      <td>0.100</td>\n",
       "      <td>0.900</td>\n",
       "      <td>1.000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>women</th>\n",
       "      <td>1.000</td>\n",
       "      <td>0.900</td>\n",
       "      <td>0.100</td>\n",
       "      <td>0.000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "pocket_binned  (4454, 5906]  (5906, 8619]  (8619, 10725]  (10725, 13103]\n",
       "menWomen                                                                \n",
       "men                   0.000         0.100          0.900           1.000\n",
       "women                 1.000         0.900          0.100           0.000"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# women's pocket sizs are smaller\n",
    "pd.crosstab(df[\"menWomen\"], df.pocket_binned, normalize=\"columns\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Data Transformations - Get women pocket size score for each brand"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "To get this score we can do a simple approach and a more complex one\n",
    "\n",
    "#### Add column is_size_greater_than_average\n",
    "Add column is_size_greater_than_average (per brand / per dataset) -> with values from 0 to 2, the number of women jeans that are larger than average in that brand"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Creating a column with data transformation of another column\n",
    "# This function is also in scripts/transformation.py\n",
    "def is_greater_than_average(series: pd.Series) -> pd.Series :\n",
    "    avg = series.mean()\n",
    "    new_series = [0 if x <= avg else 1 for x in series]\n",
    "\n",
    "    return pd.Series(new_series)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "How many pockets are of size greater than average?:  41\n"
     ]
    }
   ],
   "source": [
    "df[\"size_greater_than_average\"] = is_greater_than_average(df[\"pocketArea\"])\n",
    "print(\"How many pockets are of size greater than average?: \", df[\"size_greater_than_average\"].sum())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "41"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# the test failed, and if we uncomment the next line we see this fails too\n",
    "is_greater_than_average(df[\"pocketArea\"]).sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### TODO: Exercise 2\n",
    "\n",
    "We have accidentaly stumbled upon a situation where even though something works in the \"notebook\", when used as a function, and maybe later in a pipeline it could fail. So our test kinda failed for the wrong reasons.\n",
    "\n",
    "- Change the test to a data type test.\n",
    "- Add more tests to the function `is_greater_than_average`. How does it work when all values in the series are the same?\n",
    "\n",
    "Don't forget to add your unit tests to `test_transformation.py`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# write test"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### Add column get_pocket_size_score\n",
    "Add column get_pocket_size_score -> return a score based on if women pocket sizes are higher than average (by brand in relation to the rest maybe)\n",
    "\n",
    "\n",
    "### TODO: Exercise 3\n",
    "- add function to new script (e.g., `scoring.py`)\n",
    "- add a unit test file in test folder for the new function (e.g., `test_scoring.py`)\n",
    "- write tests. Can the following function be tested as it is written? Does it follow testable functions conventions?"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "# count number of rows with women and \"size_greater_than_average\" > 0\n",
    "# input brand, menWomen, size_greater_than_average -> 80 rows\n",
    "# output brand, score  -> less ~ 40\n",
    "\n",
    "aggr = df.groupby(by=[\"brand\", \"menWomen\"],as_index=False)[\"size_greater_than_average\"].sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "def get_sum_score_by_brand_and_gender(frame: pd.DataFrame, brand_col=\"brand\", gender_col=\"menWomen\", score_by=\"size_greater_than_average\") -> pd.DataFrame :\n",
    "    aggr = frame.groupby(by=[brand_col, gender_col],as_index=False)[score_by].sum()\n",
    "\n",
    "    return aggr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "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>brand</th>\n",
       "      <th>menWomen</th>\n",
       "      <th>size_greater_than_average</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>7 for All Mankind</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>7 for All Mankind</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Abercrombie</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Abercrombie</td>\n",
       "      <td>women</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>American Eagle</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>American Eagle</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Buckle Black</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Buckle Black</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Calvin Klein</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Calvin Klein</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Express</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Express</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Gap</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Gap</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Guess</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Guess</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>H&amp;M</td>\n",
       "      <td>men</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>H&amp;M</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>J. Crew</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>J. Crew</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>Lee</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>Lee</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>Levi's</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>Levi's</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>Lucky</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>Lucky</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>Old Navy</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>Old Navy</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>Ralph Lauren</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>Ralph Lauren</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>Tommy Hilfiger</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>Tommy Hilfiger</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>True Religion</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>True Religion</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>Uniqlo</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>Uniqlo</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>Wrangler</td>\n",
       "      <td>men</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>Wrangler</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                brand menWomen  size_greater_than_average\n",
       "0   7 for All Mankind      men                          2\n",
       "1   7 for All Mankind    women                          0\n",
       "2         Abercrombie      men                          2\n",
       "3         Abercrombie    women                          2\n",
       "4      American Eagle      men                          2\n",
       "5      American Eagle    women                          0\n",
       "6             Arizona      men                          2\n",
       "7             Arizona    women                          0\n",
       "8        Buckle Black      men                          2\n",
       "9        Buckle Black    women                          0\n",
       "10       Calvin Klein      men                          2\n",
       "11       Calvin Klein    women                          0\n",
       "12            Express      men                          2\n",
       "13            Express    women                          0\n",
       "14                Gap      men                          2\n",
       "15                Gap    women                          0\n",
       "16              Guess      men                          2\n",
       "17              Guess    women                          0\n",
       "18                H&M      men                          1\n",
       "19                H&M    women                          0\n",
       "20            J. Crew      men                          2\n",
       "21            J. Crew    women                          0\n",
       "22                Lee      men                          2\n",
       "23                Lee    women                          0\n",
       "24             Levi's      men                          2\n",
       "25             Levi's    women                          0\n",
       "26              Lucky      men                          2\n",
       "27              Lucky    women                          0\n",
       "28           Old Navy      men                          2\n",
       "29           Old Navy    women                          0\n",
       "30       Ralph Lauren      men                          2\n",
       "31       Ralph Lauren    women                          0\n",
       "32     Tommy Hilfiger      men                          2\n",
       "33     Tommy Hilfiger    women                          0\n",
       "34      True Religion      men                          2\n",
       "35      True Religion    women                          0\n",
       "36             Uniqlo      men                          2\n",
       "37             Uniqlo    women                          0\n",
       "38           Wrangler      men                          2\n",
       "39           Wrangler    women                          0"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aggr = get_sum_score_by_brand_and_gender(df, \"brand\", \"menWomen\", \"size_greater_than_average\")\n",
    "aggr"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "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>brand</th>\n",
       "      <th>menWomen</th>\n",
       "      <th>size_greater_than_average</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>7 for All Mankind</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Abercrombie</td>\n",
       "      <td>women</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>American Eagle</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Buckle Black</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Calvin Klein</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Express</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Gap</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Guess</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>H&amp;M</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>J. Crew</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>Lee</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>Levi's</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>Lucky</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>Old Navy</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>Ralph Lauren</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>Tommy Hilfiger</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>True Religion</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>Uniqlo</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>Wrangler</td>\n",
       "      <td>women</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                brand menWomen  size_greater_than_average\n",
       "1   7 for All Mankind    women                          0\n",
       "3         Abercrombie    women                          2\n",
       "5      American Eagle    women                          0\n",
       "7             Arizona    women                          0\n",
       "9        Buckle Black    women                          0\n",
       "11       Calvin Klein    women                          0\n",
       "13            Express    women                          0\n",
       "15                Gap    women                          0\n",
       "17              Guess    women                          0\n",
       "19                H&M    women                          0\n",
       "21            J. Crew    women                          0\n",
       "23                Lee    women                          0\n",
       "25             Levi's    women                          0\n",
       "27              Lucky    women                          0\n",
       "29           Old Navy    women                          0\n",
       "31       Ralph Lauren    women                          0\n",
       "33     Tommy Hilfiger    women                          0\n",
       "35      True Religion    women                          0\n",
       "37             Uniqlo    women                          0\n",
       "39           Wrangler    women                          0"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aggr[aggr.menWomen == \"women\"]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#todo: write the test"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Homework\n",
    "\n",
    "\n",
    "### a less all or nothing score\n",
    "\n",
    "we sort the pocket areas and assign to each row the index of the area value ( values should be 0 - dataframe size)\n",
    "- brand women score = average position (sum / 2)\n",
    "- brand men score = average position (sum / 2)\n",
    "- score = women / men\n",
    "\n",
    "todo: \n",
    "- write this as a testable function, add this to the `scoring.py` (you might need more than one function)\n",
    "- write unit tests for these functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_small = df[[\"brand\", \"menWomen\", \"pocketArea\"]]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_small = df_small.sort_values(\"pocketArea\").reset_index(drop=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_small['rank'] = df_small.index + 1"
   ]
  },
  {
   "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>brand</th>\n",
       "      <th>menWomen</th>\n",
       "      <th>pocketArea</th>\n",
       "      <th>rank</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Gap</td>\n",
       "      <td>women</td>\n",
       "      <td>4454.429</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Guess</td>\n",
       "      <td>women</td>\n",
       "      <td>4490.844</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Guess</td>\n",
       "      <td>women</td>\n",
       "      <td>4572.572</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Gap</td>\n",
       "      <td>women</td>\n",
       "      <td>4612.939</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Calvin Klein</td>\n",
       "      <td>women</td>\n",
       "      <td>4702.028</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>75</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>men</td>\n",
       "      <td>12667.656</td>\n",
       "      <td>76</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>76</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>men</td>\n",
       "      <td>12751.060</td>\n",
       "      <td>77</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>77</th>\n",
       "      <td>Ralph Lauren</td>\n",
       "      <td>men</td>\n",
       "      <td>12792.598</td>\n",
       "      <td>78</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>78</th>\n",
       "      <td>Wrangler</td>\n",
       "      <td>men</td>\n",
       "      <td>13079.606</td>\n",
       "      <td>79</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>79</th>\n",
       "      <td>Lee</td>\n",
       "      <td>men</td>\n",
       "      <td>13102.032</td>\n",
       "      <td>80</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>80 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           brand menWomen  pocketArea  rank\n",
       "0            Gap    women    4454.429     1\n",
       "1          Guess    women    4490.844     2\n",
       "2          Guess    women    4572.572     3\n",
       "3            Gap    women    4612.939     4\n",
       "4   Calvin Klein    women    4702.028     5\n",
       "..           ...      ...         ...   ...\n",
       "75       Arizona      men   12667.656    76\n",
       "76       Arizona      men   12751.060    77\n",
       "77  Ralph Lauren      men   12792.598    78\n",
       "78      Wrangler      men   13079.606    79\n",
       "79           Lee      men   13102.032    80\n",
       "\n",
       "[80 rows x 4 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_small"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "aggr = df_small.groupby(by=[\"brand\", \"menWomen\"],as_index=False)[\"rank\"].sum()"
   ]
  },
  {
   "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>brand</th>\n",
       "      <th>menWomen</th>\n",
       "      <th>rank</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>7 for All Mankind</td>\n",
       "      <td>men</td>\n",
       "      <td>131</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>7 for All Mankind</td>\n",
       "      <td>women</td>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Abercrombie</td>\n",
       "      <td>men</td>\n",
       "      <td>111</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Abercrombie</td>\n",
       "      <td>women</td>\n",
       "      <td>90</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>American Eagle</td>\n",
       "      <td>men</td>\n",
       "      <td>98</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>American Eagle</td>\n",
       "      <td>women</td>\n",
       "      <td>64</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>men</td>\n",
       "      <td>153</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Arizona</td>\n",
       "      <td>women</td>\n",
       "      <td>62</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Buckle Black</td>\n",
       "      <td>men</td>\n",
       "      <td>115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Buckle Black</td>\n",
       "      <td>women</td>\n",
       "      <td>57</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Calvin Klein</td>\n",
       "      <td>men</td>\n",
       "      <td>113</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Calvin Klein</td>\n",
       "      <td>women</td>\n",
       "      <td>24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Express</td>\n",
       "      <td>men</td>\n",
       "      <td>90</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Express</td>\n",
       "      <td>women</td>\n",
       "      <td>59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Gap</td>\n",
       "      <td>men</td>\n",
       "      <td>126</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Gap</td>\n",
       "      <td>women</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Guess</td>\n",
       "      <td>men</td>\n",
       "      <td>118</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Guess</td>\n",
       "      <td>women</td>\n",
       "      <td>5</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>H&amp;M</td>\n",
       "      <td>men</td>\n",
       "      <td>90</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>H&amp;M</td>\n",
       "      <td>women</td>\n",
       "      <td>55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>J. Crew</td>\n",
       "      <td>men</td>\n",
       "      <td>117</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>J. Crew</td>\n",
       "      <td>women</td>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>Lee</td>\n",
       "      <td>men</td>\n",
       "      <td>137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>Lee</td>\n",
       "      <td>women</td>\n",
       "      <td>58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>Levi's</td>\n",
       "      <td>men</td>\n",
       "      <td>124</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>Levi's</td>\n",
       "      <td>women</td>\n",
       "      <td>36</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>26</th>\n",
       "      <td>Lucky</td>\n",
       "      <td>men</td>\n",
       "      <td>115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>27</th>\n",
       "      <td>Lucky</td>\n",
       "      <td>women</td>\n",
       "      <td>44</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>Old Navy</td>\n",
       "      <td>men</td>\n",
       "      <td>129</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>29</th>\n",
       "      <td>Old Navy</td>\n",
       "      <td>women</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>Ralph Lauren</td>\n",
       "      <td>men</td>\n",
       "      <td>142</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>31</th>\n",
       "      <td>Ralph Lauren</td>\n",
       "      <td>women</td>\n",
       "      <td>21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>32</th>\n",
       "      <td>Tommy Hilfiger</td>\n",
       "      <td>men</td>\n",
       "      <td>143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>33</th>\n",
       "      <td>Tommy Hilfiger</td>\n",
       "      <td>women</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>True Religion</td>\n",
       "      <td>men</td>\n",
       "      <td>115</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>35</th>\n",
       "      <td>True Religion</td>\n",
       "      <td>women</td>\n",
       "      <td>25</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>Uniqlo</td>\n",
       "      <td>men</td>\n",
       "      <td>116</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>37</th>\n",
       "      <td>Uniqlo</td>\n",
       "      <td>women</td>\n",
       "      <td>41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>Wrangler</td>\n",
       "      <td>men</td>\n",
       "      <td>126</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>Wrangler</td>\n",
       "      <td>women</td>\n",
       "      <td>44</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                brand menWomen  rank\n",
       "0   7 for All Mankind      men   131\n",
       "1   7 for All Mankind    women    37\n",
       "2         Abercrombie      men   111\n",
       "3         Abercrombie    women    90\n",
       "4      American Eagle      men    98\n",
       "5      American Eagle    women    64\n",
       "6             Arizona      men   153\n",
       "7             Arizona    women    62\n",
       "8        Buckle Black      men   115\n",
       "9        Buckle Black    women    57\n",
       "10       Calvin Klein      men   113\n",
       "11       Calvin Klein    women    24\n",
       "12            Express      men    90\n",
       "13            Express    women    59\n",
       "14                Gap      men   126\n",
       "15                Gap    women     5\n",
       "16              Guess      men   118\n",
       "17              Guess    women     5\n",
       "18                H&M      men    90\n",
       "19                H&M    women    55\n",
       "20            J. Crew      men   117\n",
       "21            J. Crew    women    31\n",
       "22                Lee      men   137\n",
       "23                Lee    women    58\n",
       "24             Levi's      men   124\n",
       "25             Levi's    women    36\n",
       "26              Lucky      men   115\n",
       "27              Lucky    women    44\n",
       "28           Old Navy      men   129\n",
       "29           Old Navy    women    23\n",
       "30       Ralph Lauren      men   142\n",
       "31       Ralph Lauren    women    21\n",
       "32     Tommy Hilfiger      men   143\n",
       "33     Tommy Hilfiger    women    50\n",
       "34      True Religion      men   115\n",
       "35      True Religion    women    25\n",
       "36             Uniqlo      men   116\n",
       "37             Uniqlo    women    41\n",
       "38           Wrangler      men   126\n",
       "39           Wrangler    women    44"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "aggr"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# About this data\n",
    "\n",
    "source: https://github.com/the-pudding/data/tree/master/pockets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/jpeg": "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\n",
      "text/html": [
       "\n",
       "        <iframe\n",
       "            width=\"400\"\n",
       "            height=\"300\"\n",
       "            src=\"https://www.youtube.com/embed/Vi2Vgym6lbw\"\n",
       "            frameborder=\"0\"\n",
       "            allowfullscreen\n",
       "        ></iframe>\n",
       "        "
      ],
      "text/plain": [
       "<IPython.lib.display.YouTubeVideo at 0x12bee0a50>"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from IPython.lib.display import YouTubeVideo\n",
    "YouTubeVideo('Vi2Vgym6lbw')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}