03-Polars.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "a9620a09",
   "metadata": {},
   "source": [
    "# Group-By in Polars"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "53795d21",
   "metadata": {},
   "outputs": [],
   "source": [
    "import polars as pl\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import pyarrow\n",
    "\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "\n",
    "\n",
    "# Download Dataset or run in a kaggle notebook\n",
    "# https://www.kaggle.com/datasets/mkechinov/ecommerce-behavior-data-from-multi-category-store\n",
    "df = pl.read_csv(\"2019-Nov.csv\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "c795931a",
   "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",
       "\n",
       "    .dataframe td {\n",
       "        white-space: pre;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-top: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-bottom: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        line-height: 95%;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "<small>shape: (5, 9)</small>\n",
       "<thead>\n",
       "<tr>\n",
       "<th>\n",
       "event_time\n",
       "</th>\n",
       "<th>\n",
       "event_type\n",
       "</th>\n",
       "<th>\n",
       "product_id\n",
       "</th>\n",
       "<th>\n",
       "category_id\n",
       "</th>\n",
       "<th>\n",
       "category_code\n",
       "</th>\n",
       "<th>\n",
       "brand\n",
       "</th>\n",
       "<th>\n",
       "price\n",
       "</th>\n",
       "<th>\n",
       "user_id\n",
       "</th>\n",
       "<th>\n",
       "user_session\n",
       "</th>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "i64\n",
       "</td>\n",
       "<td>\n",
       "i64\n",
       "</td>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "f64\n",
       "</td>\n",
       "<td>\n",
       "i64\n",
       "</td>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "</tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;2019-11-01 00:...\n",
       "</td>\n",
       "<td>\n",
       "&quot;view&quot;\n",
       "</td>\n",
       "<td>\n",
       "1003461\n",
       "</td>\n",
       "<td>\n",
       "2053013555631882655\n",
       "</td>\n",
       "<td>\n",
       "&quot;electronics.sm...\n",
       "</td>\n",
       "<td>\n",
       "&quot;xiaomi&quot;\n",
       "</td>\n",
       "<td>\n",
       "489.07\n",
       "</td>\n",
       "<td>\n",
       "520088904\n",
       "</td>\n",
       "<td>\n",
       "&quot;4d3b30da-a5e4-...\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;2019-11-01 00:...\n",
       "</td>\n",
       "<td>\n",
       "&quot;view&quot;\n",
       "</td>\n",
       "<td>\n",
       "5000088\n",
       "</td>\n",
       "<td>\n",
       "2053013566100866035\n",
       "</td>\n",
       "<td>\n",
       "&quot;appliances.sew...\n",
       "</td>\n",
       "<td>\n",
       "&quot;janome&quot;\n",
       "</td>\n",
       "<td>\n",
       "293.65\n",
       "</td>\n",
       "<td>\n",
       "530496790\n",
       "</td>\n",
       "<td>\n",
       "&quot;8e5f4f83-366c-...\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;2019-11-01 00:...\n",
       "</td>\n",
       "<td>\n",
       "&quot;view&quot;\n",
       "</td>\n",
       "<td>\n",
       "17302664\n",
       "</td>\n",
       "<td>\n",
       "2053013553853497655\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "<td>\n",
       "&quot;creed&quot;\n",
       "</td>\n",
       "<td>\n",
       "28.31\n",
       "</td>\n",
       "<td>\n",
       "561587266\n",
       "</td>\n",
       "<td>\n",
       "&quot;755422e7-9040-...\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;2019-11-01 00:...\n",
       "</td>\n",
       "<td>\n",
       "&quot;view&quot;\n",
       "</td>\n",
       "<td>\n",
       "3601530\n",
       "</td>\n",
       "<td>\n",
       "2053013563810775923\n",
       "</td>\n",
       "<td>\n",
       "&quot;appliances.kit...\n",
       "</td>\n",
       "<td>\n",
       "&quot;lg&quot;\n",
       "</td>\n",
       "<td>\n",
       "712.87\n",
       "</td>\n",
       "<td>\n",
       "518085591\n",
       "</td>\n",
       "<td>\n",
       "&quot;3bfb58cd-7892-...\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;2019-11-01 00:...\n",
       "</td>\n",
       "<td>\n",
       "&quot;view&quot;\n",
       "</td>\n",
       "<td>\n",
       "1004775\n",
       "</td>\n",
       "<td>\n",
       "2053013555631882655\n",
       "</td>\n",
       "<td>\n",
       "&quot;electronics.sm...\n",
       "</td>\n",
       "<td>\n",
       "&quot;xiaomi&quot;\n",
       "</td>\n",
       "<td>\n",
       "183.27\n",
       "</td>\n",
       "<td>\n",
       "558856683\n",
       "</td>\n",
       "<td>\n",
       "&quot;313628f1-68b8-...\n",
       "</td>\n",
       "</tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "shape: (5, 9)\n",
       "┌──────────┬─────────────┬────────────┬───────────┬─────┬────────┬────────┬───────────┬────────────┐\n",
       "│ event_ti ┆ event_type  ┆ product_id ┆ category_ ┆ ... ┆ brand  ┆ price  ┆ user_id   ┆ user_sessi │\n",
       "│ me       ┆ ---         ┆ ---        ┆ id        ┆     ┆ ---    ┆ ---    ┆ ---       ┆ on         │\n",
       "│ ---      ┆ str         ┆ i64        ┆ ---       ┆     ┆ str    ┆ f64    ┆ i64       ┆ ---        │\n",
       "│ str      ┆             ┆            ┆ i64       ┆     ┆        ┆        ┆           ┆ str        │\n",
       "╞══════════╪═════════════╪════════════╪═══════════╪═════╪════════╪════════╪═══════════╪════════════╡\n",
       "│ 2019-11- ┆ view        ┆ 1003461    ┆ 205301355 ┆ ... ┆ xiaomi ┆ 489.07 ┆ 520088904 ┆ 4d3b30da-a │\n",
       "│ 01       ┆             ┆            ┆ 563188265 ┆     ┆        ┆        ┆           ┆ 5e4-49df-b │\n",
       "│ 00:00:00 ┆             ┆            ┆ 5         ┆     ┆        ┆        ┆           ┆ 1a8-ba5943 │\n",
       "│ UTC      ┆             ┆            ┆           ┆     ┆        ┆        ┆           ┆ f1...      │\n",
       "│ 2019-11- ┆ view        ┆ 5000088    ┆ 205301356 ┆ ... ┆ janome ┆ 293.65 ┆ 530496790 ┆ 8e5f4f83-3 │\n",
       "│ 01       ┆             ┆            ┆ 610086603 ┆     ┆        ┆        ┆           ┆ 66c-4f70-8 │\n",
       "│ 00:00:00 ┆             ┆            ┆ 5         ┆     ┆        ┆        ┆           ┆ 60e-ca7417 │\n",
       "│ UTC      ┆             ┆            ┆           ┆     ┆        ┆        ┆           ┆ 41...      │\n",
       "│ 2019-11- ┆ view        ┆ 17302664   ┆ 205301355 ┆ ... ┆ creed  ┆ 28.31  ┆ 561587266 ┆ 755422e7-9 │\n",
       "│ 01       ┆             ┆            ┆ 385349765 ┆     ┆        ┆        ┆           ┆ 040-477b-9 │\n",
       "│ 00:00:01 ┆             ┆            ┆ 5         ┆     ┆        ┆        ┆           ┆ bd2-6a6e8f │\n",
       "│ UTC      ┆             ┆            ┆           ┆     ┆        ┆        ┆           ┆ d9...      │\n",
       "│ 2019-11- ┆ view        ┆ 3601530    ┆ 205301356 ┆ ... ┆ lg     ┆ 712.87 ┆ 518085591 ┆ 3bfb58cd-7 │\n",
       "│ 01       ┆             ┆            ┆ 381077592 ┆     ┆        ┆        ┆           ┆ 892-48cc-8 │\n",
       "│ 00:00:01 ┆             ┆            ┆ 3         ┆     ┆        ┆        ┆           ┆ 020-2f17e6 │\n",
       "│ UTC      ┆             ┆            ┆           ┆     ┆        ┆        ┆           ┆ de...      │\n",
       "│ 2019-11- ┆ view        ┆ 1004775    ┆ 205301355 ┆ ... ┆ xiaomi ┆ 183.27 ┆ 558856683 ┆ 313628f1-6 │\n",
       "│ 01       ┆             ┆            ┆ 563188265 ┆     ┆        ┆        ┆           ┆ 8b8-460d-8 │\n",
       "│ 00:00:01 ┆             ┆            ┆ 5         ┆     ┆        ┆        ┆           ┆ 4f6-cec7a8 │\n",
       "│ UTC      ┆             ┆            ┆           ┆     ┆        ┆        ┆           ┆ 79...      │\n",
       "└──────────┴─────────────┴────────────┴───────────┴─────┴────────┴────────┴───────────┴────────────┘"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "d46de304",
   "metadata": {},
   "source": [
    "## 1. Groupby - pl.count (one group)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "b600993e",
   "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",
       "\n",
       "    .dataframe td {\n",
       "        white-space: pre;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-top: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-bottom: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        line-height: 95%;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "<small>shape: (5, 2)</small>\n",
       "<thead>\n",
       "<tr>\n",
       "<th>\n",
       "brand\n",
       "</th>\n",
       "<th>\n",
       "count\n",
       "</th>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "</tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;shivaki&quot;\n",
       "</td>\n",
       "<td>\n",
       "58516\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;oris&quot;\n",
       "</td>\n",
       "<td>\n",
       "6299\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;derspur&quot;\n",
       "</td>\n",
       "<td>\n",
       "1401\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;rode&quot;\n",
       "</td>\n",
       "<td>\n",
       "447\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;worth&quot;\n",
       "</td>\n",
       "<td>\n",
       "66\n",
       "</td>\n",
       "</tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "shape: (5, 2)\n",
       "┌─────────┬───────┐\n",
       "│ brand   ┆ count │\n",
       "│ ---     ┆ ---   │\n",
       "│ str     ┆ u32   │\n",
       "╞═════════╪═══════╡\n",
       "│ shivaki ┆ 58516 │\n",
       "│ oris    ┆ 6299  │\n",
       "│ derspur ┆ 1401  │\n",
       "│ rode    ┆ 447   │\n",
       "│ worth   ┆ 66    │\n",
       "└─────────┴───────┘"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby(\"brand\").agg([\n",
    "    pl.count()\n",
    "]).head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "e557dc0c",
   "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",
       "\n",
       "    .dataframe td {\n",
       "        white-space: pre;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-top: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-bottom: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        line-height: 95%;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "<small>shape: (5, 2)</small>\n",
       "<thead>\n",
       "<tr>\n",
       "<th>\n",
       "brand\n",
       "</th>\n",
       "<th>\n",
       "count\n",
       "</th>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "</tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "<td>\n",
       "9218235\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "7889245\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;apple&quot;\n",
       "</td>\n",
       "<td>\n",
       "6259379\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;xiaomi&quot;\n",
       "</td>\n",
       "<td>\n",
       "4638062\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;huawei&quot;\n",
       "</td>\n",
       "<td>\n",
       "1410126\n",
       "</td>\n",
       "</tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "shape: (5, 2)\n",
       "┌─────────┬─────────┐\n",
       "│ brand   ┆ count   │\n",
       "│ ---     ┆ ---     │\n",
       "│ str     ┆ u32     │\n",
       "╞═════════╪═════════╡\n",
       "│ null    ┆ 9218235 │\n",
       "│ samsung ┆ 7889245 │\n",
       "│ apple   ┆ 6259379 │\n",
       "│ xiaomi  ┆ 4638062 │\n",
       "│ huawei  ┆ 1410126 │\n",
       "└─────────┴─────────┘"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# sorting in polars\n",
    "df.groupby(\"brand\").agg([\n",
    "    pl.count()\n",
    "]).sort(\"count\", reverse=True).head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "eb708d83",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "(4202, 2)\n"
     ]
    },
    {
     "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>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>4109</th>\n",
       "      <td>None</td>\n",
       "      <td>9218235</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3390</th>\n",
       "      <td>samsung</td>\n",
       "      <td>7889245</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>368</th>\n",
       "      <td>apple</td>\n",
       "      <td>6259379</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3518</th>\n",
       "      <td>xiaomi</td>\n",
       "      <td>4638062</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1837</th>\n",
       "      <td>huawei</td>\n",
       "      <td>1410126</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        brand    count\n",
       "4109     None  9218235\n",
       "3390  samsung  7889245\n",
       "368     apple  6259379\n",
       "3518   xiaomi  4638062\n",
       "1837   huawei  1410126"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# aggregate in polars and convert to pandas for sorting the aggregation\n",
    "df_count = df.groupby(\"brand\").agg([\n",
    "    pl.count()\n",
    "]).to_pandas().sort_values(\"count\", ascending=False)\n",
    "\n",
    "print(df_count.shape)\n",
    "\n",
    "df_count.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "12a2a7c0",
   "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",
       "\n",
       "    .dataframe td {\n",
       "        white-space: pre;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-top: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-bottom: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        line-height: 95%;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "<small>shape: (10, 3)</small>\n",
       "<thead>\n",
       "<tr>\n",
       "<th>\n",
       "brand\n",
       "</th>\n",
       "<th>\n",
       "count\n",
       "</th>\n",
       "<th>\n",
       "n_pct\n",
       "</th>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "f64\n",
       "</td>\n",
       "</tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "<td>\n",
       "9218235\n",
       "</td>\n",
       "<td>\n",
       "0.136562\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "7889245\n",
       "</td>\n",
       "<td>\n",
       "0.116874\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;apple&quot;\n",
       "</td>\n",
       "<td>\n",
       "6259379\n",
       "</td>\n",
       "<td>\n",
       "0.092729\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;xiaomi&quot;\n",
       "</td>\n",
       "<td>\n",
       "4638062\n",
       "</td>\n",
       "<td>\n",
       "0.06871\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;huawei&quot;\n",
       "</td>\n",
       "<td>\n",
       "1410126\n",
       "</td>\n",
       "<td>\n",
       "0.02089\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;lucente&quot;\n",
       "</td>\n",
       "<td>\n",
       "1185075\n",
       "</td>\n",
       "<td>\n",
       "0.017556\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;lg&quot;\n",
       "</td>\n",
       "<td>\n",
       "1096990\n",
       "</td>\n",
       "<td>\n",
       "0.016251\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;bosch&quot;\n",
       "</td>\n",
       "<td>\n",
       "975059\n",
       "</td>\n",
       "<td>\n",
       "0.014445\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;oppo&quot;\n",
       "</td>\n",
       "<td>\n",
       "811698\n",
       "</td>\n",
       "<td>\n",
       "0.012025\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;sony&quot;\n",
       "</td>\n",
       "<td>\n",
       "798457\n",
       "</td>\n",
       "<td>\n",
       "0.011829\n",
       "</td>\n",
       "</tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "shape: (10, 3)\n",
       "┌─────────┬─────────┬──────────┐\n",
       "│ brand   ┆ count   ┆ n_pct    │\n",
       "│ ---     ┆ ---     ┆ ---      │\n",
       "│ str     ┆ u32     ┆ f64      │\n",
       "╞═════════╪═════════╪══════════╡\n",
       "│ null    ┆ 9218235 ┆ 0.136562 │\n",
       "│ samsung ┆ 7889245 ┆ 0.116874 │\n",
       "│ apple   ┆ 6259379 ┆ 0.092729 │\n",
       "│ xiaomi  ┆ 4638062 ┆ 0.06871  │\n",
       "│ ...     ┆ ...     ┆ ...      │\n",
       "│ lg      ┆ 1096990 ┆ 0.016251 │\n",
       "│ bosch   ┆ 975059  ┆ 0.014445 │\n",
       "│ oppo    ┆ 811698  ┆ 0.012025 │\n",
       "│ sony    ┆ 798457  ┆ 0.011829 │\n",
       "└─────────┴─────────┴──────────┘"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Compute count and percent of instances\n",
    "(\n",
    "    df.groupby(\"brand\").agg([\n",
    "        pl.count()\n",
    "    ])\n",
    "    .sort('count', reverse=True)\n",
    "    .with_columns([\n",
    "        (pl.col(\"count\") / pl.col(\"count\").sum()).alias(\"n_pct\")\n",
    "    ])\n",
    "    .head(10)\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "42f470ff",
   "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",
       "\n",
       "    .dataframe td {\n",
       "        white-space: pre;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-top: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-bottom: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        line-height: 95%;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "<small>shape: (5, 9)</small>\n",
       "<thead>\n",
       "<tr>\n",
       "<th>\n",
       "event_time\n",
       "</th>\n",
       "<th>\n",
       "event_type\n",
       "</th>\n",
       "<th>\n",
       "product_id\n",
       "</th>\n",
       "<th>\n",
       "category_id\n",
       "</th>\n",
       "<th>\n",
       "category_code\n",
       "</th>\n",
       "<th>\n",
       "brand\n",
       "</th>\n",
       "<th>\n",
       "price\n",
       "</th>\n",
       "<th>\n",
       "user_id\n",
       "</th>\n",
       "<th>\n",
       "user_session\n",
       "</th>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "i64\n",
       "</td>\n",
       "<td>\n",
       "i64\n",
       "</td>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "f64\n",
       "</td>\n",
       "<td>\n",
       "i64\n",
       "</td>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "</tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;2019-11-28 07:...\n",
       "</td>\n",
       "<td>\n",
       "&quot;view&quot;\n",
       "</td>\n",
       "<td>\n",
       "1005158\n",
       "</td>\n",
       "<td>\n",
       "2053013555631882655\n",
       "</td>\n",
       "<td>\n",
       "&quot;electronics.sm...\n",
       "</td>\n",
       "<td>\n",
       "&quot;xiaomi&quot;\n",
       "</td>\n",
       "<td>\n",
       "282.12\n",
       "</td>\n",
       "<td>\n",
       "544074209\n",
       "</td>\n",
       "<td>\n",
       "&quot;1ca11c10-92ac-...\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;2019-11-18 05:...\n",
       "</td>\n",
       "<td>\n",
       "&quot;view&quot;\n",
       "</td>\n",
       "<td>\n",
       "7201910\n",
       "</td>\n",
       "<td>\n",
       "2053013553165631753\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "<td>\n",
       "&quot;lego&quot;\n",
       "</td>\n",
       "<td>\n",
       "37.3\n",
       "</td>\n",
       "<td>\n",
       "541866488\n",
       "</td>\n",
       "<td>\n",
       "&quot;2ac22c2c-8d11-...\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;2019-11-01 03:...\n",
       "</td>\n",
       "<td>\n",
       "&quot;view&quot;\n",
       "</td>\n",
       "<td>\n",
       "5700384\n",
       "</td>\n",
       "<td>\n",
       "2053013553970938175\n",
       "</td>\n",
       "<td>\n",
       "&quot;auto.accessori...\n",
       "</td>\n",
       "<td>\n",
       "&quot;pioneer&quot;\n",
       "</td>\n",
       "<td>\n",
       "140.31\n",
       "</td>\n",
       "<td>\n",
       "526141854\n",
       "</td>\n",
       "<td>\n",
       "&quot;147b516f-66d1-...\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;2019-11-15 01:...\n",
       "</td>\n",
       "<td>\n",
       "&quot;view&quot;\n",
       "</td>\n",
       "<td>\n",
       "1005115\n",
       "</td>\n",
       "<td>\n",
       "2053013555631882655\n",
       "</td>\n",
       "<td>\n",
       "&quot;electronics.sm...\n",
       "</td>\n",
       "<td>\n",
       "&quot;apple&quot;\n",
       "</td>\n",
       "<td>\n",
       "914.0\n",
       "</td>\n",
       "<td>\n",
       "518773118\n",
       "</td>\n",
       "<td>\n",
       "&quot;564875fa-e866-...\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;2019-11-15 15:...\n",
       "</td>\n",
       "<td>\n",
       "&quot;view&quot;\n",
       "</td>\n",
       "<td>\n",
       "1004209\n",
       "</td>\n",
       "<td>\n",
       "2053013555631882655\n",
       "</td>\n",
       "<td>\n",
       "&quot;electronics.sm...\n",
       "</td>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "88.67\n",
       "</td>\n",
       "<td>\n",
       "529519097\n",
       "</td>\n",
       "<td>\n",
       "&quot;a19edc9b-9917-...\n",
       "</td>\n",
       "</tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "shape: (5, 9)\n",
       "┌──────────┬──────────┬──────────────┬───────────┬─────┬─────────┬────────┬───────────┬────────────┐\n",
       "│ event_ti ┆ event_ty ┆ product_id   ┆ category_ ┆ ... ┆ brand   ┆ price  ┆ user_id   ┆ user_sessi │\n",
       "│ me       ┆ pe       ┆ ---          ┆ id        ┆     ┆ ---     ┆ ---    ┆ ---       ┆ on         │\n",
       "│ ---      ┆ ---      ┆ i64          ┆ ---       ┆     ┆ str     ┆ f64    ┆ i64       ┆ ---        │\n",
       "│ str      ┆ str      ┆              ┆ i64       ┆     ┆         ┆        ┆           ┆ str        │\n",
       "╞══════════╪══════════╪══════════════╪═══════════╪═════╪═════════╪════════╪═══════════╪════════════╡\n",
       "│ 2019-11- ┆ view     ┆ 1005158      ┆ 205301355 ┆ ... ┆ xiaomi  ┆ 282.12 ┆ 544074209 ┆ 1ca11c10-9 │\n",
       "│ 28       ┆          ┆              ┆ 563188265 ┆     ┆         ┆        ┆           ┆ 2ac-4793-b │\n",
       "│ 07:26:33 ┆          ┆              ┆ 5         ┆     ┆         ┆        ┆           ┆ da6-801537 │\n",
       "│ UTC      ┆          ┆              ┆           ┆     ┆         ┆        ┆           ┆ 4f...      │\n",
       "│ 2019-11- ┆ view     ┆ 7201910      ┆ 205301355 ┆ ... ┆ lego    ┆ 37.3   ┆ 541866488 ┆ 2ac22c2c-8 │\n",
       "│ 18       ┆          ┆              ┆ 316563175 ┆     ┆         ┆        ┆           ┆ d11-41fa-a │\n",
       "│ 05:41:14 ┆          ┆              ┆ 3         ┆     ┆         ┆        ┆           ┆ e50-a7e1ef │\n",
       "│ UTC      ┆          ┆              ┆           ┆     ┆         ┆        ┆           ┆ f6...      │\n",
       "│ 2019-11- ┆ view     ┆ 5700384      ┆ 205301355 ┆ ... ┆ pioneer ┆ 140.31 ┆ 526141854 ┆ 147b516f-6 │\n",
       "│ 01       ┆          ┆              ┆ 397093817 ┆     ┆         ┆        ┆           ┆ 6d1-45d0-8 │\n",
       "│ 03:13:16 ┆          ┆              ┆ 5         ┆     ┆         ┆        ┆           ┆ a6e-2393e5 │\n",
       "│ UTC      ┆          ┆              ┆           ┆     ┆         ┆        ┆           ┆ 05...      │\n",
       "│ 2019-11- ┆ view     ┆ 1005115      ┆ 205301355 ┆ ... ┆ apple   ┆ 914.0  ┆ 518773118 ┆ 564875fa-e │\n",
       "│ 15       ┆          ┆              ┆ 563188265 ┆     ┆         ┆        ┆           ┆ 866-44d9-8 │\n",
       "│ 01:07:30 ┆          ┆              ┆ 5         ┆     ┆         ┆        ┆           ┆ 225-0338a7 │\n",
       "│ UTC      ┆          ┆              ┆           ┆     ┆         ┆        ┆           ┆ 0c...      │\n",
       "│ 2019-11- ┆ view     ┆ 1004209      ┆ 205301355 ┆ ... ┆ samsung ┆ 88.67  ┆ 529519097 ┆ a19edc9b-9 │\n",
       "│ 15       ┆          ┆              ┆ 563188265 ┆     ┆         ┆        ┆           ┆ 917-4801-8 │\n",
       "│ 15:51:08 ┆          ┆              ┆ 5         ┆     ┆         ┆        ┆           ┆ ced-c696e7 │\n",
       "│ UTC      ┆          ┆              ┆           ┆     ┆         ┆        ┆           ┆ 63...      │\n",
       "└──────────┴──────────┴──────────────┴───────────┴─────┴─────────┴────────┴───────────┴────────────┘"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.sample(5)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "c96447a2",
   "metadata": {},
   "source": [
    "### Value Counts works as expected"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "8673f115",
   "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",
       "\n",
       "    .dataframe td {\n",
       "        white-space: pre;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-top: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-bottom: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        line-height: 95%;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "<small>shape: (3, 2)</small>\n",
       "<thead>\n",
       "<tr>\n",
       "<th>\n",
       "event_type\n",
       "</th>\n",
       "<th>\n",
       "count\n",
       "</th>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "</tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;purchase&quot;\n",
       "</td>\n",
       "<td>\n",
       "916939\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;view&quot;\n",
       "</td>\n",
       "<td>\n",
       "63556110\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;cart&quot;\n",
       "</td>\n",
       "<td>\n",
       "3028930\n",
       "</td>\n",
       "</tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "shape: (3, 2)\n",
       "┌────────────┬──────────┐\n",
       "│ event_type ┆ count    │\n",
       "│ ---        ┆ ---      │\n",
       "│ str        ┆ u32      │\n",
       "╞════════════╪══════════╡\n",
       "│ purchase   ┆ 916939   │\n",
       "│ view       ┆ 63556110 │\n",
       "│ cart       ┆ 3028930  │\n",
       "└────────────┴──────────┘"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# this works!\n",
    "df['event_type'].value_counts()\n",
    "\n",
    "# More verbose version with groupby\n",
    "df.groupby(\"event_type\").agg(pl.count())"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "36ae7609",
   "metadata": {},
   "source": [
    "## 2. Groupby - pl.count (two groups)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "62cf1c49",
   "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",
       "\n",
       "    .dataframe td {\n",
       "        white-space: pre;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-top: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-bottom: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        line-height: 95%;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "<small>shape: (5, 3)</small>\n",
       "<thead>\n",
       "<tr>\n",
       "<th>\n",
       "brand\n",
       "</th>\n",
       "<th>\n",
       "event_type\n",
       "</th>\n",
       "<th>\n",
       "count\n",
       "</th>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "</tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;besty&quot;\n",
       "</td>\n",
       "<td>\n",
       "&quot;view&quot;\n",
       "</td>\n",
       "<td>\n",
       "3283\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;xgimi&quot;\n",
       "</td>\n",
       "<td>\n",
       "&quot;view&quot;\n",
       "</td>\n",
       "<td>\n",
       "1890\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;neptun&quot;\n",
       "</td>\n",
       "<td>\n",
       "&quot;view&quot;\n",
       "</td>\n",
       "<td>\n",
       "306\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;evrodetal&quot;\n",
       "</td>\n",
       "<td>\n",
       "&quot;purchase&quot;\n",
       "</td>\n",
       "<td>\n",
       "21\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;forward&quot;\n",
       "</td>\n",
       "<td>\n",
       "&quot;cart&quot;\n",
       "</td>\n",
       "<td>\n",
       "142\n",
       "</td>\n",
       "</tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "shape: (5, 3)\n",
       "┌───────────┬────────────┬───────┐\n",
       "│ brand     ┆ event_type ┆ count │\n",
       "│ ---       ┆ ---        ┆ ---   │\n",
       "│ str       ┆ str        ┆ u32   │\n",
       "╞═══════════╪════════════╪═══════╡\n",
       "│ besty     ┆ view       ┆ 3283  │\n",
       "│ xgimi     ┆ view       ┆ 1890  │\n",
       "│ neptun    ┆ view       ┆ 306   │\n",
       "│ evrodetal ┆ purchase   ┆ 21    │\n",
       "│ forward   ┆ cart       ┆ 142   │\n",
       "└───────────┴────────────┴───────┘"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_brand_event = df.groupby([\"brand\", \"event_type\"]).agg([\n",
    "    pl.count()\n",
    "])\n",
    "\n",
    "df_brand_event.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "acc7b29d",
   "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",
       "\n",
       "    .dataframe td {\n",
       "        white-space: pre;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-top: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-bottom: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        line-height: 95%;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "<small>shape: (5, 4)</small>\n",
       "<thead>\n",
       "<tr>\n",
       "<th>\n",
       "brand\n",
       "</th>\n",
       "<th>\n",
       "view\n",
       "</th>\n",
       "<th>\n",
       "purchase\n",
       "</th>\n",
       "<th>\n",
       "cart\n",
       "</th>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "</tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;besty&quot;\n",
       "</td>\n",
       "<td>\n",
       "3283\n",
       "</td>\n",
       "<td>\n",
       "24\n",
       "</td>\n",
       "<td>\n",
       "105\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;xgimi&quot;\n",
       "</td>\n",
       "<td>\n",
       "1890\n",
       "</td>\n",
       "<td>\n",
       "15\n",
       "</td>\n",
       "<td>\n",
       "66\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;neptun&quot;\n",
       "</td>\n",
       "<td>\n",
       "306\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;evrodetal&quot;\n",
       "</td>\n",
       "<td>\n",
       "7804\n",
       "</td>\n",
       "<td>\n",
       "21\n",
       "</td>\n",
       "<td>\n",
       "138\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;forward&quot;\n",
       "</td>\n",
       "<td>\n",
       "3931\n",
       "</td>\n",
       "<td>\n",
       "26\n",
       "</td>\n",
       "<td>\n",
       "142\n",
       "</td>\n",
       "</tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "shape: (5, 4)\n",
       "┌───────────┬──────┬──────────┬──────┐\n",
       "│ brand     ┆ view ┆ purchase ┆ cart │\n",
       "│ ---       ┆ ---  ┆ ---      ┆ ---  │\n",
       "│ str       ┆ u32  ┆ u32      ┆ u32  │\n",
       "╞═══════════╪══════╪══════════╪══════╡\n",
       "│ besty     ┆ 3283 ┆ 24       ┆ 105  │\n",
       "│ xgimi     ┆ 1890 ┆ 15       ┆ 66   │\n",
       "│ neptun    ┆ 306  ┆ null     ┆ null │\n",
       "│ evrodetal ┆ 7804 ┆ 21       ┆ 138  │\n",
       "│ forward   ┆ 3931 ┆ 26       ┆ 142  │\n",
       "└───────────┴──────┴──────────┴──────┘"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Convert from long to wide format\n",
    "df_brand_event_wide = df_brand_event.pivot(values=\"count\", \n",
    "                                           index=[\"brand\"], \n",
    "                                           columns=['event_type'])\n",
    "df_brand_event_wide.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "a8ab3036",
   "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",
       "\n",
       "    .dataframe td {\n",
       "        white-space: pre;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-top: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-bottom: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        line-height: 95%;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "<small>shape: (5, 4)</small>\n",
       "<thead>\n",
       "<tr>\n",
       "<th>\n",
       "brand\n",
       "</th>\n",
       "<th>\n",
       "view\n",
       "</th>\n",
       "<th>\n",
       "purchase\n",
       "</th>\n",
       "<th>\n",
       "cart\n",
       "</th>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "</tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "7091998\n",
       "</td>\n",
       "<td>\n",
       "200027\n",
       "</td>\n",
       "<td>\n",
       "597220\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;apple&quot;\n",
       "</td>\n",
       "<td>\n",
       "5603650\n",
       "</td>\n",
       "<td>\n",
       "166064\n",
       "</td>\n",
       "<td>\n",
       "489665\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "<td>\n",
       "8886720\n",
       "</td>\n",
       "<td>\n",
       "73273\n",
       "</td>\n",
       "<td>\n",
       "258242\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;xiaomi&quot;\n",
       "</td>\n",
       "<td>\n",
       "4309751\n",
       "</td>\n",
       "<td>\n",
       "68292\n",
       "</td>\n",
       "<td>\n",
       "260019\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;huawei&quot;\n",
       "</td>\n",
       "<td>\n",
       "1312663\n",
       "</td>\n",
       "<td>\n",
       "23703\n",
       "</td>\n",
       "<td>\n",
       "73760\n",
       "</td>\n",
       "</tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "shape: (5, 4)\n",
       "┌─────────┬─────────┬──────────┬────────┐\n",
       "│ brand   ┆ view    ┆ purchase ┆ cart   │\n",
       "│ ---     ┆ ---     ┆ ---      ┆ ---    │\n",
       "│ str     ┆ u32     ┆ u32      ┆ u32    │\n",
       "╞═════════╪═════════╪══════════╪════════╡\n",
       "│ samsung ┆ 7091998 ┆ 200027   ┆ 597220 │\n",
       "│ apple   ┆ 5603650 ┆ 166064   ┆ 489665 │\n",
       "│ null    ┆ 8886720 ┆ 73273    ┆ 258242 │\n",
       "│ xiaomi  ┆ 4309751 ┆ 68292    ┆ 260019 │\n",
       "│ huawei  ┆ 1312663 ┆ 23703    ┆ 73760  │\n",
       "└─────────┴─────────┴──────────┴────────┘"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Pivot and sorting\n",
    "df_brand_event_wide = ( \n",
    "    df_brand_event\n",
    "    .pivot(values=\"count\", index=[\"brand\"], columns=['event_type'])\n",
    "    .sort(\"purchase\", reverse=True)\n",
    ")\n",
    "\n",
    "df_brand_event_wide.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "e6ffbab9",
   "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",
       "\n",
       "    .dataframe td {\n",
       "        white-space: pre;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-top: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-bottom: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        line-height: 95%;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "<small>shape: (5, 7)</small>\n",
       "<thead>\n",
       "<tr>\n",
       "<th>\n",
       "brand\n",
       "</th>\n",
       "<th>\n",
       "view\n",
       "</th>\n",
       "<th>\n",
       "purchase\n",
       "</th>\n",
       "<th>\n",
       "cart\n",
       "</th>\n",
       "<th>\n",
       "cart_by_views\n",
       "</th>\n",
       "<th>\n",
       "buy_by_cart\n",
       "</th>\n",
       "<th>\n",
       "buy_by_views\n",
       "</th>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "f64\n",
       "</td>\n",
       "<td>\n",
       "f64\n",
       "</td>\n",
       "<td>\n",
       "f64\n",
       "</td>\n",
       "</tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "7091998\n",
       "</td>\n",
       "<td>\n",
       "200027\n",
       "</td>\n",
       "<td>\n",
       "597220\n",
       "</td>\n",
       "<td>\n",
       "0.08421\n",
       "</td>\n",
       "<td>\n",
       "0.33493\n",
       "</td>\n",
       "<td>\n",
       "0.028205\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;apple&quot;\n",
       "</td>\n",
       "<td>\n",
       "5603650\n",
       "</td>\n",
       "<td>\n",
       "166064\n",
       "</td>\n",
       "<td>\n",
       "489665\n",
       "</td>\n",
       "<td>\n",
       "0.087383\n",
       "</td>\n",
       "<td>\n",
       "0.339138\n",
       "</td>\n",
       "<td>\n",
       "0.029635\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "<td>\n",
       "8886720\n",
       "</td>\n",
       "<td>\n",
       "73273\n",
       "</td>\n",
       "<td>\n",
       "258242\n",
       "</td>\n",
       "<td>\n",
       "0.029059\n",
       "</td>\n",
       "<td>\n",
       "0.283738\n",
       "</td>\n",
       "<td>\n",
       "0.008245\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;xiaomi&quot;\n",
       "</td>\n",
       "<td>\n",
       "4309751\n",
       "</td>\n",
       "<td>\n",
       "68292\n",
       "</td>\n",
       "<td>\n",
       "260019\n",
       "</td>\n",
       "<td>\n",
       "0.060333\n",
       "</td>\n",
       "<td>\n",
       "0.262642\n",
       "</td>\n",
       "<td>\n",
       "0.015846\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "&quot;huawei&quot;\n",
       "</td>\n",
       "<td>\n",
       "1312663\n",
       "</td>\n",
       "<td>\n",
       "23703\n",
       "</td>\n",
       "<td>\n",
       "73760\n",
       "</td>\n",
       "<td>\n",
       "0.056191\n",
       "</td>\n",
       "<td>\n",
       "0.321353\n",
       "</td>\n",
       "<td>\n",
       "0.018057\n",
       "</td>\n",
       "</tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "shape: (5, 7)\n",
       "┌─────────┬─────────┬──────────┬────────┬───────────────┬─────────────┬──────────────┐\n",
       "│ brand   ┆ view    ┆ purchase ┆ cart   ┆ cart_by_views ┆ buy_by_cart ┆ buy_by_views │\n",
       "│ ---     ┆ ---     ┆ ---      ┆ ---    ┆ ---           ┆ ---         ┆ ---          │\n",
       "│ str     ┆ u32     ┆ u32      ┆ u32    ┆ f64           ┆ f64         ┆ f64          │\n",
       "╞═════════╪═════════╪══════════╪════════╪═══════════════╪═════════════╪══════════════╡\n",
       "│ samsung ┆ 7091998 ┆ 200027   ┆ 597220 ┆ 0.08421       ┆ 0.33493     ┆ 0.028205     │\n",
       "│ apple   ┆ 5603650 ┆ 166064   ┆ 489665 ┆ 0.087383      ┆ 0.339138    ┆ 0.029635     │\n",
       "│ null    ┆ 8886720 ┆ 73273    ┆ 258242 ┆ 0.029059      ┆ 0.283738    ┆ 0.008245     │\n",
       "│ xiaomi  ┆ 4309751 ┆ 68292    ┆ 260019 ┆ 0.060333      ┆ 0.262642    ┆ 0.015846     │\n",
       "│ huawei  ┆ 1312663 ┆ 23703    ┆ 73760  ┆ 0.056191      ┆ 0.321353    ┆ 0.018057     │\n",
       "└─────────┴─────────┴──────────┴────────┴───────────────┴─────────────┴──────────────┘"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Keep original columns and compute % values\n",
    "agg_performance = df_brand_event_wide.with_columns([\n",
    "    (pl.col(\"cart\") / pl.col(\"view\")).alias(\"cart_by_views\"),\n",
    "    (pl.col(\"purchase\") / pl.col(\"cart\")).alias(\"buy_by_cart\"),\n",
    "    (pl.col(\"purchase\") / pl.col(\"view\")).alias(\"buy_by_views\"),\n",
    "])\n",
    "agg_performance.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "f449f47d",
   "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>cart_by_views</th>\n",
       "      <th>buy_by_cart</th>\n",
       "      <th>buy_by_views</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0.00</th>\n",
       "      <td>0.001115</td>\n",
       "      <td>0.026810</td>\n",
       "      <td>0.000353</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.25</th>\n",
       "      <td>0.014640</td>\n",
       "      <td>0.210685</td>\n",
       "      <td>0.003704</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.50</th>\n",
       "      <td>0.024419</td>\n",
       "      <td>0.279579</td>\n",
       "      <td>0.006616</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.75</th>\n",
       "      <td>0.037220</td>\n",
       "      <td>0.343597</td>\n",
       "      <td>0.010874</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0.99</th>\n",
       "      <td>0.121112</td>\n",
       "      <td>0.750000</td>\n",
       "      <td>0.043861</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1.00</th>\n",
       "      <td>0.400000</td>\n",
       "      <td>0.900000</td>\n",
       "      <td>0.200000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      cart_by_views  buy_by_cart  buy_by_views\n",
       "0.00       0.001115     0.026810      0.000353\n",
       "0.25       0.014640     0.210685      0.003704\n",
       "0.50       0.024419     0.279579      0.006616\n",
       "0.75       0.037220     0.343597      0.010874\n",
       "0.99       0.121112     0.750000      0.043861\n",
       "1.00       0.400000     0.900000      0.200000"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "agg_performance.filter(pl.col(\"buy_by_cart\") < 1)\\\n",
    ".select(['cart_by_views', 'buy_by_cart', 'buy_by_views'])\\\n",
    ".to_pandas().quantile([0, 0.25, 0.5, 0.75, 0.99, 1])\n",
    "\n",
    "# .boxplot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "9a85fd1c",
   "metadata": {},
   "outputs": [],
   "source": [
    "df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "462114ea",
   "metadata": {},
   "source": [
    "## 3. Group By - UserId + 2 Groups"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "91d715c1",
   "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",
       "\n",
       "    .dataframe td {\n",
       "        white-space: pre;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-top: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-bottom: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        line-height: 95%;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "<small>shape: (5, 4)</small>\n",
       "<thead>\n",
       "<tr>\n",
       "<th>\n",
       "user_id\n",
       "</th>\n",
       "<th>\n",
       "brand\n",
       "</th>\n",
       "<th>\n",
       "event_type\n",
       "</th>\n",
       "<th>\n",
       "count\n",
       "</th>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "i64\n",
       "</td>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "</tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr>\n",
       "<td>\n",
       "552111808\n",
       "</td>\n",
       "<td>\n",
       "&quot;apple&quot;\n",
       "</td>\n",
       "<td>\n",
       "&quot;view&quot;\n",
       "</td>\n",
       "<td>\n",
       "2\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "538399224\n",
       "</td>\n",
       "<td>\n",
       "&quot;nokia&quot;\n",
       "</td>\n",
       "<td>\n",
       "&quot;view&quot;\n",
       "</td>\n",
       "<td>\n",
       "1\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "554511060\n",
       "</td>\n",
       "<td>\n",
       "&quot;bosch&quot;\n",
       "</td>\n",
       "<td>\n",
       "&quot;view&quot;\n",
       "</td>\n",
       "<td>\n",
       "6\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "515375111\n",
       "</td>\n",
       "<td>\n",
       "&quot;payot&quot;\n",
       "</td>\n",
       "<td>\n",
       "&quot;view&quot;\n",
       "</td>\n",
       "<td>\n",
       "1\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "563928304\n",
       "</td>\n",
       "<td>\n",
       "&quot;apple&quot;\n",
       "</td>\n",
       "<td>\n",
       "&quot;cart&quot;\n",
       "</td>\n",
       "<td>\n",
       "2\n",
       "</td>\n",
       "</tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "shape: (5, 4)\n",
       "┌───────────┬───────┬────────────┬───────┐\n",
       "│ user_id   ┆ brand ┆ event_type ┆ count │\n",
       "│ ---       ┆ ---   ┆ ---        ┆ ---   │\n",
       "│ i64       ┆ str   ┆ str        ┆ u32   │\n",
       "╞═══════════╪═══════╪════════════╪═══════╡\n",
       "│ 552111808 ┆ apple ┆ view       ┆ 2     │\n",
       "│ 538399224 ┆ nokia ┆ view       ┆ 1     │\n",
       "│ 554511060 ┆ bosch ┆ view       ┆ 6     │\n",
       "│ 515375111 ┆ payot ┆ view       ┆ 1     │\n",
       "│ 563928304 ┆ apple ┆ cart       ┆ 2     │\n",
       "└───────────┴───────┴────────────┴───────┘"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# groupby \n",
    "# %%timeit\n",
    "# 8.72 s ± 246 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)\n",
    "\n",
    "df_user_brand_event = df.groupby([\"user_id\", \"brand\", \"event_type\"]).agg([\n",
    "    pl.count()\n",
    "])\n",
    "\n",
    "df_user_brand_event.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "ef226232",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(17798395, 4)"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_user_brand_event.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "333cec58",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(15905970, 5)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# pivot and sort\n",
    "df_user_brand_event = (df_user_brand_event\n",
    "    .pivot(values=\"count\", index=[\"user_id\", \"brand\"], columns=['event_type'])\n",
    "    .sort(\"purchase\", reverse=True)\n",
    ")\n",
    "df_user_brand_event.shape"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "b49f2faa",
   "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",
       "\n",
       "    .dataframe td {\n",
       "        white-space: pre;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-top: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-bottom: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        line-height: 95%;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "<small>shape: (5, 5)</small>\n",
       "<thead>\n",
       "<tr>\n",
       "<th>\n",
       "user_id\n",
       "</th>\n",
       "<th>\n",
       "brand\n",
       "</th>\n",
       "<th>\n",
       "view\n",
       "</th>\n",
       "<th>\n",
       "cart\n",
       "</th>\n",
       "<th>\n",
       "purchase\n",
       "</th>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "i64\n",
       "</td>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "</tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr>\n",
       "<td>\n",
       "564068124\n",
       "</td>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "634\n",
       "</td>\n",
       "<td>\n",
       "431\n",
       "</td>\n",
       "<td>\n",
       "453\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "518514099\n",
       "</td>\n",
       "<td>\n",
       "&quot;apple&quot;\n",
       "</td>\n",
       "<td>\n",
       "314\n",
       "</td>\n",
       "<td>\n",
       "206\n",
       "</td>\n",
       "<td>\n",
       "190\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "549030056\n",
       "</td>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "313\n",
       "</td>\n",
       "<td>\n",
       "300\n",
       "</td>\n",
       "<td>\n",
       "180\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "543128872\n",
       "</td>\n",
       "<td>\n",
       "&quot;apple&quot;\n",
       "</td>\n",
       "<td>\n",
       "239\n",
       "</td>\n",
       "<td>\n",
       "161\n",
       "</td>\n",
       "<td>\n",
       "154\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "549109608\n",
       "</td>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "307\n",
       "</td>\n",
       "<td>\n",
       "184\n",
       "</td>\n",
       "<td>\n",
       "154\n",
       "</td>\n",
       "</tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "shape: (5, 5)\n",
       "┌───────────┬─────────┬──────┬──────┬──────────┐\n",
       "│ user_id   ┆ brand   ┆ view ┆ cart ┆ purchase │\n",
       "│ ---       ┆ ---     ┆ ---  ┆ ---  ┆ ---      │\n",
       "│ i64       ┆ str     ┆ u32  ┆ u32  ┆ u32      │\n",
       "╞═══════════╪═════════╪══════╪══════╪══════════╡\n",
       "│ 564068124 ┆ samsung ┆ 634  ┆ 431  ┆ 453      │\n",
       "│ 518514099 ┆ apple   ┆ 314  ┆ 206  ┆ 190      │\n",
       "│ 549030056 ┆ samsung ┆ 313  ┆ 300  ┆ 180      │\n",
       "│ 543128872 ┆ apple   ┆ 239  ┆ 161  ┆ 154      │\n",
       "│ 549109608 ┆ samsung ┆ 307  ┆ 184  ┆ 154      │\n",
       "└───────────┴─────────┴──────┴──────┴──────────┘"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_user_brand_event.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "4970c9b9",
   "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",
       "\n",
       "    .dataframe td {\n",
       "        white-space: pre;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-top: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-bottom: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        line-height: 95%;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "<small>shape: (10, 5)</small>\n",
       "<thead>\n",
       "<tr>\n",
       "<th>\n",
       "user_id\n",
       "</th>\n",
       "<th>\n",
       "brand\n",
       "</th>\n",
       "<th>\n",
       "view\n",
       "</th>\n",
       "<th>\n",
       "cart\n",
       "</th>\n",
       "<th>\n",
       "purchase\n",
       "</th>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "i64\n",
       "</td>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "</tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr>\n",
       "<td>\n",
       "515799072\n",
       "</td>\n",
       "<td>\n",
       "&quot;vitek&quot;\n",
       "</td>\n",
       "<td>\n",
       "2\n",
       "</td>\n",
       "<td>\n",
       "1\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "516363897\n",
       "</td>\n",
       "<td>\n",
       "&quot;xiaomi&quot;\n",
       "</td>\n",
       "<td>\n",
       "55\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "578708569\n",
       "</td>\n",
       "<td>\n",
       "&quot;giorgioarmani&quot;\n",
       "</td>\n",
       "<td>\n",
       "1\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "514489669\n",
       "</td>\n",
       "<td>\n",
       "&quot;oneplus&quot;\n",
       "</td>\n",
       "<td>\n",
       "2\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "555238866\n",
       "</td>\n",
       "<td>\n",
       "&quot;willmark&quot;\n",
       "</td>\n",
       "<td>\n",
       "1\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "517923794\n",
       "</td>\n",
       "<td>\n",
       "&quot;jbl&quot;\n",
       "</td>\n",
       "<td>\n",
       "3\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "549669920\n",
       "</td>\n",
       "<td>\n",
       "&quot;haier&quot;\n",
       "</td>\n",
       "<td>\n",
       "1\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "518042667\n",
       "</td>\n",
       "<td>\n",
       "&quot;casio&quot;\n",
       "</td>\n",
       "<td>\n",
       "2\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "567725178\n",
       "</td>\n",
       "<td>\n",
       "&quot;rondell&quot;\n",
       "</td>\n",
       "<td>\n",
       "5\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "559073920\n",
       "</td>\n",
       "<td>\n",
       "&quot;element&quot;\n",
       "</td>\n",
       "<td>\n",
       "8\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "</tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "shape: (10, 5)\n",
       "┌───────────┬───────────────┬──────┬──────┬──────────┐\n",
       "│ user_id   ┆ brand         ┆ view ┆ cart ┆ purchase │\n",
       "│ ---       ┆ ---           ┆ ---  ┆ ---  ┆ ---      │\n",
       "│ i64       ┆ str           ┆ u32  ┆ u32  ┆ u32      │\n",
       "╞═══════════╪═══════════════╪══════╪══════╪══════════╡\n",
       "│ 515799072 ┆ vitek         ┆ 2    ┆ 1    ┆ null     │\n",
       "│ 516363897 ┆ xiaomi        ┆ 55   ┆ null ┆ null     │\n",
       "│ 578708569 ┆ giorgioarmani ┆ 1    ┆ null ┆ null     │\n",
       "│ 514489669 ┆ oneplus       ┆ 2    ┆ null ┆ null     │\n",
       "│ ...       ┆ ...           ┆ ...  ┆ ...  ┆ ...      │\n",
       "│ 549669920 ┆ haier         ┆ 1    ┆ null ┆ null     │\n",
       "│ 518042667 ┆ casio         ┆ 2    ┆ null ┆ null     │\n",
       "│ 567725178 ┆ rondell       ┆ 5    ┆ null ┆ null     │\n",
       "│ 559073920 ┆ element       ┆ 8    ┆ null ┆ null     │\n",
       "└───────────┴───────────────┴──────┴──────┴──────────┘"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_user_brand_event.sample(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "11487bcf",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_user_brand_event = df_user_brand_event.with_columns([\n",
    "    pl.col('purchase').fill_null(strategy=\"zero\"),\n",
    "    pl.col('view').fill_null(strategy=\"zero\"),\n",
    "    pl.col('cart').fill_null(strategy=\"zero\"),\n",
    "])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "a08c8623",
   "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",
       "\n",
       "    .dataframe td {\n",
       "        white-space: pre;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-top: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-bottom: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        line-height: 95%;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "<small>shape: (10, 5)</small>\n",
       "<thead>\n",
       "<tr>\n",
       "<th>\n",
       "user_id\n",
       "</th>\n",
       "<th>\n",
       "brand\n",
       "</th>\n",
       "<th>\n",
       "view\n",
       "</th>\n",
       "<th>\n",
       "cart\n",
       "</th>\n",
       "<th>\n",
       "purchase\n",
       "</th>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "i64\n",
       "</td>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "</tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr>\n",
       "<td>\n",
       "541978247\n",
       "</td>\n",
       "<td>\n",
       "&quot;lanvin&quot;\n",
       "</td>\n",
       "<td>\n",
       "1\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "516056866\n",
       "</td>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "3\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "551017434\n",
       "</td>\n",
       "<td>\n",
       "&quot;rowenta&quot;\n",
       "</td>\n",
       "<td>\n",
       "1\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "579916513\n",
       "</td>\n",
       "<td>\n",
       "&quot;apple&quot;\n",
       "</td>\n",
       "<td>\n",
       "2\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "561227529\n",
       "</td>\n",
       "<td>\n",
       "&quot;bosch&quot;\n",
       "</td>\n",
       "<td>\n",
       "6\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "549667503\n",
       "</td>\n",
       "<td>\n",
       "&quot;arg&quot;\n",
       "</td>\n",
       "<td>\n",
       "1\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "521552552\n",
       "</td>\n",
       "<td>\n",
       "&quot;philips&quot;\n",
       "</td>\n",
       "<td>\n",
       "1\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "513035989\n",
       "</td>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "2\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "512869754\n",
       "</td>\n",
       "<td>\n",
       "&quot;maribel&quot;\n",
       "</td>\n",
       "<td>\n",
       "14\n",
       "</td>\n",
       "<td>\n",
       "1\n",
       "</td>\n",
       "<td>\n",
       "1\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "570056543\n",
       "</td>\n",
       "<td>\n",
       "&quot;greyder&quot;\n",
       "</td>\n",
       "<td>\n",
       "1\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "</tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "shape: (10, 5)\n",
       "┌───────────┬─────────┬──────┬──────┬──────────┐\n",
       "│ user_id   ┆ brand   ┆ view ┆ cart ┆ purchase │\n",
       "│ ---       ┆ ---     ┆ ---  ┆ ---  ┆ ---      │\n",
       "│ i64       ┆ str     ┆ u32  ┆ u32  ┆ u32      │\n",
       "╞═══════════╪═════════╪══════╪══════╪══════════╡\n",
       "│ 541978247 ┆ lanvin  ┆ 1    ┆ 0    ┆ 0        │\n",
       "│ 516056866 ┆ samsung ┆ 3    ┆ 0    ┆ 0        │\n",
       "│ 551017434 ┆ rowenta ┆ 1    ┆ 0    ┆ 0        │\n",
       "│ 579916513 ┆ apple   ┆ 2    ┆ 0    ┆ 0        │\n",
       "│ ...       ┆ ...     ┆ ...  ┆ ...  ┆ ...      │\n",
       "│ 521552552 ┆ philips ┆ 1    ┆ 0    ┆ 0        │\n",
       "│ 513035989 ┆ samsung ┆ 2    ┆ 0    ┆ 0        │\n",
       "│ 512869754 ┆ maribel ┆ 14   ┆ 1    ┆ 1        │\n",
       "│ 570056543 ┆ greyder ┆ 1    ┆ 0    ┆ 0        │\n",
       "└───────────┴─────────┴──────┴──────┴──────────┘"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_user_brand_event.sample(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "9c64d449",
   "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",
       "\n",
       "    .dataframe td {\n",
       "        white-space: pre;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-top: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-bottom: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        line-height: 95%;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "<small>shape: (10, 5)</small>\n",
       "<thead>\n",
       "<tr>\n",
       "<th>\n",
       "user_id\n",
       "</th>\n",
       "<th>\n",
       "brand\n",
       "</th>\n",
       "<th>\n",
       "view\n",
       "</th>\n",
       "<th>\n",
       "cart\n",
       "</th>\n",
       "<th>\n",
       "purchase\n",
       "</th>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "i64\n",
       "</td>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "</tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr>\n",
       "<td>\n",
       "564068124\n",
       "</td>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "634\n",
       "</td>\n",
       "<td>\n",
       "431\n",
       "</td>\n",
       "<td>\n",
       "453\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "518514099\n",
       "</td>\n",
       "<td>\n",
       "&quot;apple&quot;\n",
       "</td>\n",
       "<td>\n",
       "314\n",
       "</td>\n",
       "<td>\n",
       "206\n",
       "</td>\n",
       "<td>\n",
       "190\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "549030056\n",
       "</td>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "313\n",
       "</td>\n",
       "<td>\n",
       "300\n",
       "</td>\n",
       "<td>\n",
       "180\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "543128872\n",
       "</td>\n",
       "<td>\n",
       "&quot;apple&quot;\n",
       "</td>\n",
       "<td>\n",
       "239\n",
       "</td>\n",
       "<td>\n",
       "161\n",
       "</td>\n",
       "<td>\n",
       "154\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "549109608\n",
       "</td>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "307\n",
       "</td>\n",
       "<td>\n",
       "184\n",
       "</td>\n",
       "<td>\n",
       "154\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "567928887\n",
       "</td>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "174\n",
       "</td>\n",
       "<td>\n",
       "217\n",
       "</td>\n",
       "<td>\n",
       "118\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "521230795\n",
       "</td>\n",
       "<td>\n",
       "&quot;apple&quot;\n",
       "</td>\n",
       "<td>\n",
       "315\n",
       "</td>\n",
       "<td>\n",
       "195\n",
       "</td>\n",
       "<td>\n",
       "118\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "513230794\n",
       "</td>\n",
       "<td>\n",
       "&quot;lg&quot;\n",
       "</td>\n",
       "<td>\n",
       "334\n",
       "</td>\n",
       "<td>\n",
       "102\n",
       "</td>\n",
       "<td>\n",
       "116\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "543312954\n",
       "</td>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "302\n",
       "</td>\n",
       "<td>\n",
       "129\n",
       "</td>\n",
       "<td>\n",
       "115\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "538473314\n",
       "</td>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "639\n",
       "</td>\n",
       "<td>\n",
       "178\n",
       "</td>\n",
       "<td>\n",
       "114\n",
       "</td>\n",
       "</tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "shape: (10, 5)\n",
       "┌───────────┬─────────┬──────┬──────┬──────────┐\n",
       "│ user_id   ┆ brand   ┆ view ┆ cart ┆ purchase │\n",
       "│ ---       ┆ ---     ┆ ---  ┆ ---  ┆ ---      │\n",
       "│ i64       ┆ str     ┆ u32  ┆ u32  ┆ u32      │\n",
       "╞═══════════╪═════════╪══════╪══════╪══════════╡\n",
       "│ 564068124 ┆ samsung ┆ 634  ┆ 431  ┆ 453      │\n",
       "│ 518514099 ┆ apple   ┆ 314  ┆ 206  ┆ 190      │\n",
       "│ 549030056 ┆ samsung ┆ 313  ┆ 300  ┆ 180      │\n",
       "│ 543128872 ┆ apple   ┆ 239  ┆ 161  ┆ 154      │\n",
       "│ ...       ┆ ...     ┆ ...  ┆ ...  ┆ ...      │\n",
       "│ 521230795 ┆ apple   ┆ 315  ┆ 195  ┆ 118      │\n",
       "│ 513230794 ┆ lg      ┆ 334  ┆ 102  ┆ 116      │\n",
       "│ 543312954 ┆ samsung ┆ 302  ┆ 129  ┆ 115      │\n",
       "│ 538473314 ┆ samsung ┆ 639  ┆ 178  ┆ 114      │\n",
       "└───────────┴─────────┴──────┴──────┴──────────┘"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Looks like these user-ids are distributors (nobody need 400 samsung devices)\n",
    "df_user_brand_event.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "8828efb6",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_user_brand_event = df_user_brand_event.with_columns(\n",
    "    (pl.col(\"purchase\") / pl.col(\"view\")).alias(\"pct_buy_views\")\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "297c8ee9",
   "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",
       "\n",
       "    .dataframe td {\n",
       "        white-space: pre;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-top: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-bottom: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        line-height: 95%;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "<small>shape: (5, 6)</small>\n",
       "<thead>\n",
       "<tr>\n",
       "<th>\n",
       "user_id\n",
       "</th>\n",
       "<th>\n",
       "brand\n",
       "</th>\n",
       "<th>\n",
       "view\n",
       "</th>\n",
       "<th>\n",
       "cart\n",
       "</th>\n",
       "<th>\n",
       "purchase\n",
       "</th>\n",
       "<th>\n",
       "pct_buy_views\n",
       "</th>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "i64\n",
       "</td>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "f64\n",
       "</td>\n",
       "</tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr>\n",
       "<td>\n",
       "564068124\n",
       "</td>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "634\n",
       "</td>\n",
       "<td>\n",
       "431\n",
       "</td>\n",
       "<td>\n",
       "453\n",
       "</td>\n",
       "<td>\n",
       "0.714511\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "518514099\n",
       "</td>\n",
       "<td>\n",
       "&quot;apple&quot;\n",
       "</td>\n",
       "<td>\n",
       "314\n",
       "</td>\n",
       "<td>\n",
       "206\n",
       "</td>\n",
       "<td>\n",
       "190\n",
       "</td>\n",
       "<td>\n",
       "0.605096\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "549030056\n",
       "</td>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "313\n",
       "</td>\n",
       "<td>\n",
       "300\n",
       "</td>\n",
       "<td>\n",
       "180\n",
       "</td>\n",
       "<td>\n",
       "0.57508\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "543128872\n",
       "</td>\n",
       "<td>\n",
       "&quot;apple&quot;\n",
       "</td>\n",
       "<td>\n",
       "239\n",
       "</td>\n",
       "<td>\n",
       "161\n",
       "</td>\n",
       "<td>\n",
       "154\n",
       "</td>\n",
       "<td>\n",
       "0.644351\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "549109608\n",
       "</td>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "307\n",
       "</td>\n",
       "<td>\n",
       "184\n",
       "</td>\n",
       "<td>\n",
       "154\n",
       "</td>\n",
       "<td>\n",
       "0.501629\n",
       "</td>\n",
       "</tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "shape: (5, 6)\n",
       "┌───────────┬─────────┬──────┬──────┬──────────┬───────────────┐\n",
       "│ user_id   ┆ brand   ┆ view ┆ cart ┆ purchase ┆ pct_buy_views │\n",
       "│ ---       ┆ ---     ┆ ---  ┆ ---  ┆ ---      ┆ ---           │\n",
       "│ i64       ┆ str     ┆ u32  ┆ u32  ┆ u32      ┆ f64           │\n",
       "╞═══════════╪═════════╪══════╪══════╪══════════╪═══════════════╡\n",
       "│ 564068124 ┆ samsung ┆ 634  ┆ 431  ┆ 453      ┆ 0.714511      │\n",
       "│ 518514099 ┆ apple   ┆ 314  ┆ 206  ┆ 190      ┆ 0.605096      │\n",
       "│ 549030056 ┆ samsung ┆ 313  ┆ 300  ┆ 180      ┆ 0.57508       │\n",
       "│ 543128872 ┆ apple   ┆ 239  ┆ 161  ┆ 154      ┆ 0.644351      │\n",
       "│ 549109608 ┆ samsung ┆ 307  ┆ 184  ┆ 154      ┆ 0.501629      │\n",
       "└───────────┴─────────┴──────┴──────┴──────────┴───────────────┘"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_user_brand_event.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "id": "8dcedc31",
   "metadata": {},
   "outputs": [],
   "source": [
    "df_user_brand_event = df_user_brand_event.with_columns(\n",
    "    pl.when(pl.col(\"pct_buy_views\").is_infinite())\n",
    "      .then(None)\n",
    "      .otherwise(pl.col(\"pct_buy_views\"))\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "id": "f8717498",
   "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",
       "\n",
       "    .dataframe td {\n",
       "        white-space: pre;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-top: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        padding-bottom: 0;\n",
       "    }\n",
       "\n",
       "    .dataframe td {\n",
       "        line-height: 95%;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "<small>shape: (5, 7)</small>\n",
       "<thead>\n",
       "<tr>\n",
       "<th>\n",
       "user_id\n",
       "</th>\n",
       "<th>\n",
       "brand\n",
       "</th>\n",
       "<th>\n",
       "view\n",
       "</th>\n",
       "<th>\n",
       "cart\n",
       "</th>\n",
       "<th>\n",
       "purchase\n",
       "</th>\n",
       "<th>\n",
       "pct_buy_views\n",
       "</th>\n",
       "<th>\n",
       "literal\n",
       "</th>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "i64\n",
       "</td>\n",
       "<td>\n",
       "str\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "u32\n",
       "</td>\n",
       "<td>\n",
       "f64\n",
       "</td>\n",
       "<td>\n",
       "f64\n",
       "</td>\n",
       "</tr>\n",
       "</thead>\n",
       "<tbody>\n",
       "<tr>\n",
       "<td>\n",
       "523994890\n",
       "</td>\n",
       "<td>\n",
       "&quot;chicco&quot;\n",
       "</td>\n",
       "<td>\n",
       "22\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0.0\n",
       "</td>\n",
       "<td>\n",
       "0.0\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "544424277\n",
       "</td>\n",
       "<td>\n",
       "&quot;samsung&quot;\n",
       "</td>\n",
       "<td>\n",
       "1\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0.0\n",
       "</td>\n",
       "<td>\n",
       "0.0\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "566059853\n",
       "</td>\n",
       "<td>\n",
       "&quot;huawei&quot;\n",
       "</td>\n",
       "<td>\n",
       "1\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0.0\n",
       "</td>\n",
       "<td>\n",
       "0.0\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "524404734\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "<td>\n",
       "8\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0.0\n",
       "</td>\n",
       "<td>\n",
       "0.0\n",
       "</td>\n",
       "</tr>\n",
       "<tr>\n",
       "<td>\n",
       "556587564\n",
       "</td>\n",
       "<td>\n",
       "null\n",
       "</td>\n",
       "<td>\n",
       "2\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0\n",
       "</td>\n",
       "<td>\n",
       "0.0\n",
       "</td>\n",
       "<td>\n",
       "0.0\n",
       "</td>\n",
       "</tr>\n",
       "</tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "shape: (5, 7)\n",
       "┌───────────┬─────────┬──────┬──────┬──────────┬───────────────┬─────────┐\n",
       "│ user_id   ┆ brand   ┆ view ┆ cart ┆ purchase ┆ pct_buy_views ┆ literal │\n",
       "│ ---       ┆ ---     ┆ ---  ┆ ---  ┆ ---      ┆ ---           ┆ ---     │\n",
       "│ i64       ┆ str     ┆ u32  ┆ u32  ┆ u32      ┆ f64           ┆ f64     │\n",
       "╞═══════════╪═════════╪══════╪══════╪══════════╪═══════════════╪═════════╡\n",
       "│ 523994890 ┆ chicco  ┆ 22   ┆ 0    ┆ 0        ┆ 0.0           ┆ 0.0     │\n",
       "│ 544424277 ┆ samsung ┆ 1    ┆ 0    ┆ 0        ┆ 0.0           ┆ 0.0     │\n",
       "│ 566059853 ┆ huawei  ┆ 1    ┆ 0    ┆ 0        ┆ 0.0           ┆ 0.0     │\n",
       "│ 524404734 ┆ null    ┆ 8    ┆ 0    ┆ 0        ┆ 0.0           ┆ 0.0     │\n",
       "│ 556587564 ┆ null    ┆ 2    ┆ 0    ┆ 0        ┆ 0.0           ┆ 0.0     │\n",
       "└───────────┴─────────┴──────┴──────┴──────────┴───────────────┴─────────┘"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_user_brand_event.sample(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "dd65472f",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "4d94fcdc",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "a6e69b67",
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "80f5e219",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python [conda env:tabular]",
   "language": "python",
   "name": "conda-env-tabular-py"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.9.16"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}