FULL COURSE TASKS.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#FULL COURSE SQL LANGUAGE LESSONS "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "#  Clickhouse (with Redash & Tabix) was used to solve the tasks."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# LESSON 1 -- Intro to SQL"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Show the first 10 rows.\n",
    "SELECT \n",
    "  * \n",
    "FROM \n",
    "  checks \n",
    "LIMIT \n",
    "  10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Show top-10 values in Rub column. Sort them in descending order.\n",
    "SELECT \n",
    "  * \n",
    "FROM \n",
    "  checks \n",
    "ORDER BY \n",
    "  Rub DESC \n",
    "LIMIT \n",
    "  10;"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Show the first 15 rows of the table renaming Rub to Revenue \n",
    "# and sort the result by UserID (ascending order).\n",
    "SELECT \n",
    "  Rub AS Revenue, \n",
    "  BuyDate, \n",
    "  UserID \n",
    "FROM \n",
    "  checks \n",
    "ORDER BY \n",
    "  UserID ASC \n",
    "LIMIT \n",
    "  15"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Check the dates range we have in the checks table.\n",
    "SELECT \n",
    "  MIN(BuyDate) AS MinDate, \n",
    "  MAX(BuyDate) AS MaxDate \n",
    "FROM \n",
    "  checks"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Show 10 unique UserID's. Sort the results by UserID in ascending order.\n",
    "SELECT \n",
    "  DISTINCT UserID \n",
    "FROM \n",
    "  checks \n",
    "ORDER BY \n",
    "  UserID ASC \n",
    "LIMIT \n",
    "  10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Show the sales data: orders purchased on the 8th of March 2019.\n",
    "# Sort the results by Rub in descending order. Limit the data to the first 10 rows.\n",
    "SELECT \n",
    "  * \n",
    "FROM \n",
    "  checks \n",
    "WHERE \n",
    "  BuyDate IN ('2019-03-08') \n",
    "ORDER BY \n",
    "  Rub DESC \n",
    "LIMIT \n",
    "  10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Show unique UserID's for buyers made purchase on the 1st of September 2019\n",
    "# and their expenditures were higer than 2000 rubles.\n",
    "# Sort the resul by UserID in descending order.\n",
    "SELECT \n",
    "  DISTINCT UserID \n",
    "FROM \n",
    "  checks \n",
    "WHERE \n",
    "  (\n",
    "    BuyDate = '2019-09-01' \n",
    "    AND Rub > 2000\n",
    "  ) \n",
    "ORDER BY \n",
    "  UserID DESC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# LESSON 2 -- Grouping and Aggregate"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Count number of purchases per each client, name the column as NumChecks.\n",
    "# Sort the result by NumChecks in descending order. Limit output to 10 rows.\n",
    "SELECT \n",
    "  UserID, \n",
    "  COUNT(Rub) AS NumChecks \n",
    "FROM \n",
    "  checks \n",
    "GROUP BY \n",
    "  UserID \n",
    "ORDER BY \n",
    "  NumChecks DESC \n",
    "LIMIT \n",
    "  10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Count sum of purchases made by each client, name the column as Revenue.\n",
    "# Sort the result by Revenue in descending order. Limit output to 10 rows.\n",
    "SELECT \n",
    "  UserID, \n",
    "  COUNT(Rub) AS NumChecks, \n",
    "  SUM(Rub) AS Revenue \n",
    "FROM \n",
    "  checks \n",
    "GROUP BY \n",
    "  UserID \n",
    "ORDER BY \n",
    "  Revenue DESC \n",
    "LIMIT \n",
    "  10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Count by date min, max and average money spent.\n",
    "# Use sort by BuyDate in descending order. Limit output ot 10 rows.\n",
    "SELECT \n",
    "  BuyDate, \n",
    "  MIN(Rub) AS MinCheck, \n",
    "  MAX(Rub) AS MaxCheck, \n",
    "  AVG(Rub) AS AvgCheck \n",
    "FROM \n",
    "  checks \n",
    "GROUP BY \n",
    "  BuyDate \n",
    "ORDER BY \n",
    "  BuyDate DESC \n",
    "LIMIT \n",
    "  10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Find customers who has spent more than 10000 rubles.\n",
    "# Sort the result by UserID in descending order. Limit output to 10 rows.\n",
    "SELECT \n",
    "  UserID, \n",
    "  SUM(Rub) AS Revenue \n",
    "FROM \n",
    "  checks \n",
    "GROUP BY \n",
    "  UserID \n",
    "HAVING \n",
    "  Revenue > 10000 \n",
    "ORDER BY \n",
    "  UserID DESC \n",
    "LIMIT \n",
    "  10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Count Revenue by country. Sort the result by Revenue in descenig order.\n",
    "# Limit output to 10 rows.\n",
    "SELECT \n",
    "  Country, \n",
    "  SUM(Quantity * UnitPrice) AS Revenue \n",
    "FROM \n",
    "  default.retail \n",
    "GROUP BY \n",
    "  Country \n",
    "ORDER BY \n",
    "  Revenue DESC \n",
    "LIMIT \n",
    "  10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Count average number of purchases and an average price of a unit.\n",
    "# Group the result by country and sort it by an average price of a unit \n",
    "# in descending order. Do not include purchases having \"Manual\" in Description.\n",
    "SELECT \n",
    "  Country, \n",
    "  AVG(UnitPrice), \n",
    "  AVG(Quantity) \n",
    "FROM \n",
    "  default.retail \n",
    "WHERE \n",
    "  Description != 'Manual' \n",
    "GROUP BY \n",
    "  Country \n",
    "ORDER BY \n",
    "  AVG(UnitPrice) DESC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate Revenue by month. Sort the result by Revenue in descending order.\n",
    "# Do not include purchases having \"Manual\" in Description.\n",
    "SELECT \n",
    "  toStartOfMonth(InvoiceDate) AS month, \n",
    "  SUM(UnitPrice * Quantity) AS revenue \n",
    "FROM \n",
    "  default.retail \n",
    "WHERE \n",
    "  Description != 'Manual' \n",
    "GROUP BY \n",
    "  month \n",
    "ORDER BY \n",
    "  revenue DESC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate an average price of a unit bought in march 2011.\n",
    "# Group the results by CustomerID and sort it by an average price of a unit in\n",
    "# descending order. Do not include purchases having \"Manual\" in Description.\n",
    "# Limit output to 10 rows.\n",
    "SELECT \n",
    "  CustomerID, \n",
    "  AVG(UnitPrice) AS avg_purch_in_march \n",
    "FROM \n",
    "  default.retail \n",
    "WHERE \n",
    "  Description != 'Manual' \n",
    "GROUP BY \n",
    "  CustomerID \n",
    "HAVING \n",
    "  toStartOfMonth(InvoiceDate) = '2011-03-01' \n",
    "ORDER BY \n",
    "  avg_purch_in_march DESC \n",
    "LIMIT \n",
    "  10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Group the data by month and country and calculate average, min, max quanitity of\n",
    "# bought items. Sort the result by average quantity in descending order.\n",
    "# Do not include purchases having \"Manual\" in Description and purchases having a\n",
    "# negative quantity. Show the results for United Kingdom only.\n",
    "SELECT \n",
    "  Country, \n",
    "  toStartOfMonth(InvoiceDate) AS month, \n",
    "  AVG(Quantity) AS avg_quantity, \n",
    "  MIN(Quantity) AS min_quantity, \n",
    "  MAX(Quantity) AS max_quantity \n",
    "FROM \n",
    "  default.retail \n",
    "WHERE \n",
    "  Description != 'Manual' \n",
    "  AND Quantity > 0 \n",
    "GROUP BY \n",
    "  Country, \n",
    "  month \n",
    "HAVING \n",
    "  Country = 'United Kingdom' \n",
    "ORDER BY \n",
    "  avg_quantity DESC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# LESSON 3 -- Joins"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Join two tables on DeviceID. Sort the result by DeviceID in descenig order.\n",
    "# Limit output to 100 rows.\n",
    "SELECT \n",
    "  l.AppPlatform AS AppPlatform, \n",
    "  l.events AS events, \n",
    "  l.EventDate AS EventDate, \n",
    "  r.UserID AS UserID \n",
    "FROM \n",
    "  events AS l \n",
    "  LEFT JOIN devices AS r ON l.DeviceID = r.DeviceID \n",
    "ORDER BY \n",
    "  DeviceID DESC \n",
    "LIMIT \n",
    "  100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Join three tables to find out the Source that brough users with the most amount \n",
    "# of money spent. Sort the result by money spent in descenig order. \n",
    "# Limit output to 100 rows.\n",
    "SELECT \n",
    "  l.Source, \n",
    "  SUM(r.Rub) AS Rub \n",
    "FROM \n",
    "  (\n",
    "    SELECT \n",
    "      l.Source, \n",
    "      l.DeviceID, \n",
    "      r.UserID \n",
    "    FROM \n",
    "      installs AS l \n",
    "      JOIN devices AS r ON l.DeviceID = r.DeviceID\n",
    "  ) AS l \n",
    "  JOIN checks AS r ON l.UserID = r.UserID \n",
    "GROUP BY \n",
    "  l.Source \n",
    "ORDER BY \n",
    "  Rub DESC \n",
    "LIMIT \n",
    "  100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Find the number of unique users who bought something in our app.\n",
    "# Group and sort the result by Source. Limit output to 100 rows.\n",
    "SELECT \n",
    "  l.Source, \n",
    "  uniqExact(r.UserID) \n",
    "FROM \n",
    "  (\n",
    "    SELECT \n",
    "      l.Source, \n",
    "      l.DeviceID, \n",
    "      r.UserID \n",
    "    FROM \n",
    "      installs AS l \n",
    "      JOIN devices AS r ON l.DeviceID = r.DeviceID\n",
    "  ) AS l \n",
    "  JOIN checks AS r ON l.UserID = r.UserID \n",
    "GROUP BY \n",
    "  l.Source \n",
    "ORDER BY \n",
    "  l.Source DESC \n",
    "LIMIT \n",
    "  100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate Revenue, min, max, average check.\n",
    "# Group and sort the result by Source. Limit output to 100 rows.\n",
    "SELECT \n",
    "  l.Source, \n",
    "  SUM(r.Rub) AS Revenue, \n",
    "  MIN(r.Rub) AS min_check, \n",
    "  MAX(r.Rub) AS max_check, \n",
    "  AVG(r.Rub) AS avg_check \n",
    "FROM \n",
    "  (\n",
    "    SELECT \n",
    "      l.Source, \n",
    "      l.DeviceID, \n",
    "      r.UserID \n",
    "    FROM \n",
    "      installs AS l \n",
    "      JOIN devices AS r ON l.DeviceID = r.DeviceID\n",
    "  ) AS l \n",
    "  JOIN checks AS r ON l.UserID = r.UserID \n",
    "GROUP BY \n",
    "  l.Source \n",
    "ORDER BY \n",
    "  l.Source DESC \n",
    "LIMIT \n",
    "  100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Show DeviceID's of users who has made at least one purchase over the last month\n",
    "# (October 2019). Sort the result by DeviceID in ascending order.\n",
    "# Limit output to 100 rows.\n",
    "SELECT \n",
    "  r.BuyDate, \n",
    "  l.DeviceID, \n",
    "  SUM(r.Rub) AS total \n",
    "FROM \n",
    "  (\n",
    "    SELECT \n",
    "      l.Source, \n",
    "      l.DeviceID, \n",
    "      r.UserID \n",
    "    FROM \n",
    "      installs AS l \n",
    "      JOIN devices AS r ON l.DeviceID = r.DeviceID\n",
    "  ) AS l \n",
    "  JOIN checks AS r ON l.UserID = r.UserID \n",
    "GROUP BY \n",
    "  l.DeviceID, \n",
    "  r.BuyDate \n",
    "HAVING \n",
    "  total > 0 \n",
    "  AND toStartOfMonth(\n",
    "    CAST(r.BuyDate AS Date)\n",
    "  ) = '2019-10-01' \n",
    "ORDER BY \n",
    "  l.DeviceID ASC \n",
    "LIMIT \n",
    "  100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Check an average amount of views (events) by users from various Platforms and\n",
    "# Sources. Sort the result by an average amount of views for each platform-source pair\n",
    "# in descending order. Limit output to 100 rows.\n",
    "SELECT \n",
    "  r.Platform, \n",
    "  r.Source, \n",
    "  AVG(l.events) AS avg_views \n",
    "FROM \n",
    "  events AS l \n",
    "  JOIN installs AS r ON l.DeviceID = r.DeviceID \n",
    "GROUP BY \n",
    "  r.Platform, \n",
    "  r.Source \n",
    "ORDER BY \n",
    "  avg_views DESC \n",
    "LIMIT \n",
    "  100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Count the number of unique DeviceIDs in installs only for those DeviceIDs that\n",
    "# have views (events) in events table for Platform Android.\n",
    "SELECT \n",
    "  COUNT(DISTINCT r.DeviceID) \n",
    "FROM \n",
    "  events AS l \n",
    "  INNER JOIN installs AS r ON l.DeviceID = r.DeviceID \n",
    "WHERE \n",
    "  r.Platform = 'android' \n",
    "LIMIT \n",
    "  100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate the conversion rate from installs to views for users using iOS platform.\n",
    "SELECT \n",
    "  (\n",
    "    COUNT(DISTINCT r.DeviceID) / COUNT(DISTINCT l.DeviceID)\n",
    "  ) AS conversion \n",
    "FROM \n",
    "  installs AS l FULL \n",
    "  JOIN events AS r ON l.DeviceID = r.DeviceID \n",
    "WHERE \n",
    "  l.Platform = 'iOS' "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Print 10 unique DeviceIDs that are presented in the table events but are missing from the table\n",
    "# installs because of a logging mistake. List them by DeviceIDs in descending order.\n",
    "  DISTINCT(r.DeviceID) \n",
    "FROM \n",
    "  installs AS l RIGHT ANTI \n",
    "  JOIN events AS r ON l.DeviceID = r.DeviceID \n",
    "ORDER BY \n",
    "  r.DeviceID DESC \n",
    "LIMIT \n",
    "  10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# LESSON 4 -- Data Types"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Find the month and the year with the most number of new hosts registartions\n",
    "SELECT \n",
    "  COUNT(DISTINCT host_id) AS hosts_num, \n",
    "  toStartOfMonth(\n",
    "    toDateOrNull(host_since)\n",
    "  ) AS start_date \n",
    "FROM \n",
    "  default.listings \n",
    "WHERE \n",
    "  start_date IS NOT NULL \n",
    "GROUP BY \n",
    "  start_date \n",
    "ORDER BY \n",
    "  hosts_num DESC \n",
    "Limit \n",
    "  100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Find an average response rate for hosts and superhosts\n",
    "\n",
    "# for superhosts\n",
    "SELECT \n",
    "  AVG(sup.freq) AS avg_response_super \n",
    "FROM \n",
    "  (\n",
    "    SELECT \n",
    "      DISTINCT host_id, \n",
    "      toInt8OrNull(\n",
    "        replaceAll(host_response_rate, '%', '')\n",
    "      ) AS freq, \n",
    "      host_is_superhost \n",
    "    FROM \n",
    "      default.listings \n",
    "    WHERE \n",
    "      host_is_superhost == 't' \n",
    "      AND freq IS NOT NULL\n",
    "  ) AS sup\n",
    "\n",
    "# for regular hosts\n",
    "SELECT \n",
    "  AVG(not_sup.freq) AS avg_response_regular \n",
    "FROM \n",
    "  (\n",
    "    SELECT \n",
    "      DISTINCT host_id, \n",
    "      toInt8OrNull(\n",
    "        replaceAll(host_response_rate, '%', '')\n",
    "      ) AS freq, \n",
    "      host_is_superhost \n",
    "    FROM \n",
    "      default.listings \n",
    "    WHERE \n",
    "      host_is_superhost == 'f' \n",
    "      AND freq IS NOT NULL\n",
    "  ) AS not_sup"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Alternative solution using only one query\n",
    "SELECT \n",
    "  host_is_superhost, \n",
    "  AVG(\n",
    "    toInt8OrNull(\n",
    "      replaceAll(host_response_rate, '%', '')\n",
    "    )\n",
    "  ) AS avg_response_rate \n",
    "FROM \n",
    "  (\n",
    "    SELECT \n",
    "      host_is_superhost, \n",
    "      host_response_rate, \n",
    "      host_id \n",
    "    FROM \n",
    "      default.listings \n",
    "    GROUP BY \n",
    "      host_is_superhost, \n",
    "      host_response_rate, \n",
    "      host_id\n",
    "  ) \n",
    "GROUP BY \n",
    "  host_is_superhost"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Assign property ids to arrays and list them grouped by host_id.\n",
    "# Caclculate an average price of a property for a host_id. \n",
    "# Find the host_id with the highest average price. \n",
    "SELECT \n",
    "  DISTINCT host_id, \n",
    "  AVG(\n",
    "    toDecimal32OrNull(\n",
    "      replaceRegexpAll(price, '[$,]', ''), \n",
    "      2\n",
    "    )\n",
    "  ) AS avg_price, \n",
    "  groupArray(id) \n",
    "FROM \n",
    "  default.listings \n",
    "GROUP BY \n",
    "  host_id \n",
    "ORDER BY \n",
    "  avg_price DESC, \n",
    "  host_id DESC \n",
    "LIMIT \n",
    "  100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate the difference between the min and max price for each host_id\n",
    "SELECT \n",
    "  DISTINCT host_id, \n",
    "  MIN(\n",
    "    toDecimal32OrNull(\n",
    "      replaceRegexpAll(price, '[$,]', ''), \n",
    "      2\n",
    "    )\n",
    "  ) AS min_price, \n",
    "  MAX(\n",
    "    toDecimal32OrNull(\n",
    "      replaceRegexpAll(price, '[$,]', ''), \n",
    "      2\n",
    "    )\n",
    "  ) AS max_price, \n",
    "  max_price - min_price AS diff \n",
    "FROM \n",
    "  default.listings \n",
    "GROUP BY \n",
    "  host_id \n",
    "HAVING \n",
    "  diff > 0 \n",
    "ORDER BY \n",
    "  diff DESC, \n",
    "  host_id DESC \n",
    "LIMIT \n",
    "  100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Now group the data by a room_type.\n",
    "# Calculate an average price, average security deposit and an average cleaning fee.\n",
    "SELECT \n",
    "  room_type, \n",
    "  AVG(\n",
    "    toDecimal32OrNull(\n",
    "      replaceRegexpAll(price, '[$,]', ''), \n",
    "      2\n",
    "    )\n",
    "  ) AS avg_price, \n",
    "  AVG(\n",
    "    toDecimal32OrNull(\n",
    "      replaceRegexpAll(security_deposit, '[$,]', ''), \n",
    "      2\n",
    "    )\n",
    "  ) AS avg_deposit, \n",
    "  AVG(\n",
    "    toDecimal32OrNull(\n",
    "      replaceRegexpAll(cleaning_fee, '[$,]', ''), \n",
    "      2\n",
    "    )\n",
    "  ) AS avg_cleaning \n",
    "FROM \n",
    "  default.listings \n",
    "GROUP BY \n",
    "  room_type \n",
    "ORDER BY \n",
    "  avg_deposit DESC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# What districts do have the lowest average price per night?\n",
    "SELECT \n",
    "  neighbourhood_cleansed, \n",
    "  AVG(\n",
    "    toDecimal32OrNull(\n",
    "      replaceRegexpAll(price, '[$,]', ''), \n",
    "      2\n",
    "    )\n",
    "  ) AS avg_price \n",
    "FROM \n",
    "  default.listings \n",
    "GROUP BY \n",
    "  neighbourhood_cleansed \n",
    "ORDER BY \n",
    "  avg_price ASC \n",
    "LIMIT \n",
    "  100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# What districts do have the largest average square of property for a room type\n",
    "# 'Entire home/apt'. Sort by average squre and list the top-3.\n",
    "SELECT \n",
    "  neighbourhood_cleansed, \n",
    "  AVG(\n",
    "    toDecimal32OrNull(square_feet, 2)\n",
    "  ) AS avg_square \n",
    "FROM \n",
    "  default.listings \n",
    "WHERE \n",
    "  room_type == 'Entire home/apt' \n",
    "GROUP BY \n",
    "  neighbourhood_cleansed \n",
    "ORDER BY \n",
    "  avg_square DESC \n",
    "LIMIT \n",
    "  3"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Find the 'Private room' having the closest location to the \n",
    "# city center (13.4050, 52.5200)\n",
    "SELECT \n",
    "  id, \n",
    "  toFloat64OrNull(latitude) AS lat, \n",
    "  toFloat64OrNull(longitude) AS lng, \n",
    "  geoDistance(13.4050, 52.5200, lng, lat) AS dist \n",
    "FROM \n",
    "  default.listings \n",
    "WHERE \n",
    "  room_type == 'Private room' \n",
    "ORDER BY \n",
    "  dist ASC \n",
    "Limit \n",
    "  100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# LESSON 5 -- WITH, VIEWS and Creating Tables"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# PART 1"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Show advertisement where review score is above average and \n",
    "# the number of reviews per month is less than three.\n",
    "# Sort the result in descending order by number of reviews per month and \n",
    "# by review score.\n",
    "WITH (\n",
    "  SELECT \n",
    "    AVG(\n",
    "      toDecimal32OrNull(review_scores_rating, 2)\n",
    "    ) \n",
    "  FROM \n",
    "    listings \n",
    "  WHERE \n",
    "    toDecimal32OrNull(review_scores_rating, 2) IS NOT NULL\n",
    ") AS avg_scores_rating\n",
    "\n",
    "\n",
    "SELECT \n",
    "  id, \n",
    "  toDecimal32OrNull(review_scores_rating, 2) AS scores_rating, \n",
    "  avg_scores_rating, \n",
    "  toDecimal32(reviews_per_month, 2) AS review_per_month \n",
    "FROM \n",
    "  listings \n",
    "WHERE \n",
    "  scores_rating > avg_scores_rating \n",
    "  AND review_per_month < 3 \n",
    "ORDER BY \n",
    "  review_per_month DESC, \n",
    "  scores_rating DESC \n",
    "LIMIT \n",
    "  100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate an average distance to the city center and show host_ids\n",
    "# where property type is 'Private room' and the distance to the city center\n",
    "# is less than an average distance to the city center.\n",
    "# Pick a room that is further from the city center but is located closer \n",
    "# than an average distance to the city center.\n",
    "WITH (\n",
    "  SELECT \n",
    "    AVG(\n",
    "      geoDistance(\n",
    "        13.4050, \n",
    "        52.5200, \n",
    "        toFloat64OrNull(longitude), \n",
    "        toFloat64OrNull(latitude)\n",
    "      )\n",
    "    ) \n",
    "  FROM \n",
    "    listings \n",
    "  WHERE \n",
    "    room_type == 'Private room'\n",
    ") AS avg_dist \n",
    "\n",
    "\n",
    "SELECT \n",
    "  host_id, \n",
    "  toFloat64OrNull(latitude) AS lat, \n",
    "  toFloat64OrNull(longitude) AS lng, \n",
    "  geoDistance(13.4050, 52.5200, lng, lat) AS dist, \n",
    "  avg_dist \n",
    "FROM \n",
    "  default.listings \n",
    "WHERE \n",
    "  room_type == 'Private room' \n",
    "  AND dist < avg_dist \n",
    "ORDER BY \n",
    "  dist DESC \n",
    "Limit \n",
    "  100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Query advertisements where distance to the city center is less than\n",
    "# an averaga distance to the city center, price per night including a \n",
    "# cleaning fee is less than 100 USD, having the last review date not later\n",
    "# than '2018-09-01' and have wifi in the list of amenities.\n",
    "# Sort the result by review_scores_rating in descending order.\n",
    "WITH (\n",
    "  SELECT \n",
    "    AVG(\n",
    "      geoDistance(\n",
    "        13.4050, \n",
    "        52.5200, \n",
    "        toFloat64OrNull(longitude), \n",
    "        toFloat64OrNull(latitude)\n",
    "      )\n",
    "    ) \n",
    "  FROM \n",
    "    listings\n",
    ") AS avg_dist \n",
    "\n",
    "\n",
    "SELECT \n",
    "  host_id, \n",
    "  toDecimal32OrNull(\n",
    "    replaceRegexpAll(price, '[$,]', ''), \n",
    "    2\n",
    "  ) * 7 + toDecimal32OrNull(\n",
    "    replaceRegexpAll(cleaning_fee, '[$,]', ''), \n",
    "    2\n",
    "  ) AS weekly_price, \n",
    "  cleaning_fee, \n",
    "  price, \n",
    "  weekly_price / 7 AS daily_price, \n",
    "  geoDistance(\n",
    "    13.4050, \n",
    "    52.5200, \n",
    "    toFloat64OrNull(longitude), \n",
    "    toFloat64OrNull(latitude)\n",
    "  ) AS dist, \n",
    "  avg_dist, \n",
    "  toFloat32OrNull(review_scores_rating) AS scores_rating, \n",
    "  last_review, \n",
    "  amenities \n",
    "FROM \n",
    "  default.listings \n",
    "WHERE \n",
    "  (\n",
    "    dist < avg_dist \n",
    "    AND multiSearchAnyCaseInsensitive(amenities, [ 'wifi' ]) != 0\n",
    "  ) \n",
    "  AND (\n",
    "    daily_price < 100 \n",
    "    AND last_review > '2018-08-31'\n",
    "  ) \n",
    "ORDER BY \n",
    "  scores_rating DESC \n",
    "LIMIT \n",
    "  100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Pick available advertisements having number of reviews from unique users \n",
    "# above average. Sort the result by listing_id in ascending order.\n",
    "WITH (\n",
    "  SELECT \n",
    "    AVG(reviewer_id) \n",
    "  FROM \n",
    "    (\n",
    "      SELECT \n",
    "        COUNT(\n",
    "          DISTINCT(reviewer_id)\n",
    "        ) AS reviewer_id, \n",
    "        listing_id \n",
    "      FROM \n",
    "        reviews \n",
    "      GROUP BY \n",
    "        listing_id\n",
    "    )\n",
    ") AS avg_uniq_revs \n",
    "\n",
    "\n",
    "SELECT \n",
    "  reviewer_id, \n",
    "  listing_id \n",
    "FROM \n",
    "  (\n",
    "    SELECT \n",
    "      COUNT(\n",
    "        DISTINCT(reviewer_id)\n",
    "      ) AS reviewer_id, \n",
    "      listing_id \n",
    "    FROM \n",
    "      reviews \n",
    "    GROUP BY \n",
    "      listing_id\n",
    "  ) AS a \n",
    "  JOIN (\n",
    "    SELECT \n",
    "      listing_id \n",
    "    FROM \n",
    "      calendar_summary \n",
    "    WHERE \n",
    "      available == 't' \n",
    "    GROUP BY \n",
    "      listing_id\n",
    "  ) AS b ON a.listing_id = b.listing_id \n",
    "WHERE \n",
    "  reviewer_id > avg_uniq_revs \n",
    "ORDER BY \n",
    "  listing_id ASC \n",
    "LIMIT \n",
    "  100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Split customers by four segments.\n",
    "SELECT \n",
    "  CASE WHEN AVG(Rub) < 5 THEN 'A' WHEN AVG(Rub) >= 5 \n",
    "  AND AVG(Rub) < 10 THEN 'B' WHEN AVG(Rub) >= 10 \n",
    "  AND AVG(Rub) < 20 THEN 'C' ELSE 'D' END AS level, \n",
    "  UserID, \n",
    "  AVG(Rub) AS avg_rub \n",
    "FROM \n",
    "  checks \n",
    "GROUP BY \n",
    "  UserID \n",
    "ORDER BY \n",
    "  UserID ASC \n",
    "LIMIT \n",
    "  4"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Find a number of customers for each segment and calculate\n",
    "# the revenue for each segment. Sort the result by revenue in a\n",
    "# descending order.\n",
    "SELECT \n",
    "  COUNT(DISTINCT UserID) AS num_of_clients, \n",
    "  level, \n",
    "  SUM(sum_rub) AS revenue \n",
    "FROM \n",
    "  (\n",
    "    SELECT \n",
    "      CASE WHEN AVG(Rub) < 5 THEN 'A' WHEN AVG(Rub) >= 5 \n",
    "      AND AVG(Rub) < 10 THEN 'B' WHEN AVG(Rub) >= 10 \n",
    "      AND AVG(Rub) < 20 THEN 'C' ELSE 'D' END AS level, \n",
    "      UserID, \n",
    "      AVG(Rub) AS avg_rub, \n",
    "      SUM(Rub) AS sum_rub \n",
    "    FROM \n",
    "      checks \n",
    "    GROUP BY \n",
    "      UserID \n",
    "    ORDER BY \n",
    "      UserID ASC\n",
    "  ) \n",
    "GROUP BY \n",
    "  level \n",
    "ORDER BY \n",
    "  revenue DESC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# If the listed property has a kitchen and a flexible cancellation policy\n",
    "# rate it as 'good'; if it has only kithcen rate it as 'ok'; if it has neither \n",
    "# a kithcen nor a flexible cancallation policy rate it as 'not ok'.\n",
    "# Sort the result by the new column with your rating.\n",
    "SELECT \n",
    "  host_id, \n",
    "  CASE WHEN multiSearchAnyCaseInsensitive(amenities, [ 'kitchen' ]) != 0 \n",
    "  AND multiSearchAnyCaseInsensitive(\n",
    "    cancellation_policy, [ 'flexible' ]\n",
    "  ) != 0 THEN 'good' WHEN multiSearchAnyCaseInsensitive(amenities, [ 'kitchen' ]) != 0 \n",
    "  AND multiSearchAnyCaseInsensitive(\n",
    "    cancellation_policy, [ 'flexible' ]\n",
    "  ) == 0 THEN 'ok' ELSE 'not ok' END my_group \n",
    "FROM \n",
    "  default.listings \n",
    "ORDER BY \n",
    "  my_group ASC \n",
    "LIMIT \n",
    "  5"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# PART 2"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a new table having columns with a proper data type.\n",
    "CREATE TABLE test.reviews (\n",
    "  listing_id UInt32, \n",
    "  id UInt32, \n",
    "  date DateTime('Europe/Moscow'), \n",
    "  reviewer_id UInt32, \n",
    "  reviewer_name String, \n",
    "  comments String\n",
    ") ENGINE = MergeTree \n",
    "ORDER BY \n",
    "  listing_id, \n",
    "  id"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Change the data type for date from DateTime to Date.\n",
    "ALTER TABLE \n",
    "  test.reviews MODIFY COLUMN date Date"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Delete rows with empty comments from reviews table.\n",
    "ALTER TABLE \n",
    "  test.reviews DELETE \n",
    "WHERE \n",
    "  comments == ''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a view having reviewer_id and a number of ids from a reviews table.\n",
    "CREATE VIEW test.reviews AS (\n",
    "  SELECT \n",
    "    reviewer_id, \n",
    "    COUNT(id) AS reviews_count, \n",
    "  FROM \n",
    "    test.reviews \n",
    "  GROUP BY \n",
    "    reviewer_id\n",
    ")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a new column reviewer_score (UInt8) in a reviews table \n",
    "# right after a reviewer_name column.\n",
    "ALTER TABLE \n",
    "  test.reviews \n",
    "ADD \n",
    "  COLUMN reviewer_score UInt8 \n",
    "AFTER \n",
    "  reviewer_name"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a new column price (Float32) in a reviews table \n",
    "# right after a comments column.\n",
    "ALTER TABLE \n",
    "  test.reviews \n",
    "ADD \n",
    "  COLUMN price Float32 \n",
    "AFTER \n",
    "  comments"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Update the price column by doubling its values for all lines\n",
    "# having date > '2019-01-01'\n",
    "ALTER TABLE \n",
    "  test.reviews \n",
    "UPDATE \n",
    "  price = price * 2 \n",
    "WHERE \n",
    "  date > '2019-01-01'"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# LESSON 6 -- ETL (Python + SQL)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# to get data from database in Clickhouse we'll use Pandahouse library "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# importing libraries\n",
    "import pandahouse as ph\n",
    "import pandas as pd\n",
    "import seaborn as sns\n",
    "import matplotlib.pyplot as plt\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "# creating connection to Clickhouse\n",
    "connection = dict(database='default',\n",
    "                  host='https://clickhouse.lab.karpov.courses',\n",
    "                  user='login',  # an actual login was removed\n",
    "                  password='password')  # an actual password was removed"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Check that you've connected to Clickhouse by showing the number of rows in the database**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>COUNT()</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>85207</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "   COUNT()\n",
       "0    85207"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# querying data from Clickhouse\n",
    "query = \"\"\"\n",
    "SELECT\n",
    "COUNT()\n",
    "FROM default.ldn_listings\n",
    "LIMIT 1000\n",
    "\"\"\"\n",
    "df = ph.read_clickhouse(query, connection=connection)\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Import to a dataframe price per night depending on the booked property type. Limit the query to 1000 lines.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "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>id</th>\n",
       "      <th>room_type</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>13913</td>\n",
       "      <td>Private room</td>\n",
       "      <td>65.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>15400</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>17402</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>300.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>17506</td>\n",
       "      <td>Private room</td>\n",
       "      <td>150.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>25023</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>65.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>995</th>\n",
       "      <td>541081</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>110.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>996</th>\n",
       "      <td>541363</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>110.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>997</th>\n",
       "      <td>542231</td>\n",
       "      <td>Private room</td>\n",
       "      <td>65.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>998</th>\n",
       "      <td>542646</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>999</th>\n",
       "      <td>542754</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>80.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         id        room_type  price\n",
       "0     13913     Private room   65.0\n",
       "1     15400  Entire home/apt  100.0\n",
       "2     17402  Entire home/apt  300.0\n",
       "3     17506     Private room  150.0\n",
       "4     25023  Entire home/apt   65.0\n",
       "..      ...              ...    ...\n",
       "995  541081  Entire home/apt  110.0\n",
       "996  541363  Entire home/apt  110.0\n",
       "997  542231     Private room   65.0\n",
       "998  542646  Entire home/apt  100.0\n",
       "999  542754  Entire home/apt   80.0\n",
       "\n",
       "[1000 rows x 3 columns]"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# querying data from Clickhouse\n",
    "query = \"\"\"\n",
    "SELECT\n",
    "    id,\n",
    "    room_type,\n",
    "    toFloat32OrNull(replaceRegexpAll(price, '[$,]', '')) AS price\n",
    "FROM default.ldn_listings\n",
    "ORDER BY id\n",
    "LIMIT 1000\n",
    "\"\"\"\n",
    "df = ph.read_clickhouse(query, connection=connection)\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Group the dataframe by a booked property type and find the 75 quantile of a stay per night.**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "room_type\n",
       "Entire home/apt    160.00\n",
       "Hotel room         157.00\n",
       "Private room        70.00\n",
       "Shared room         69.25\n",
       "Name: price, dtype: float64"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.groupby('room_type').price.quantile(0.75)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Add review scores rating data to our dataframe**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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>id</th>\n",
       "      <th>room_type</th>\n",
       "      <th>price</th>\n",
       "      <th>review_scores_rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>13913</td>\n",
       "      <td>Private room</td>\n",
       "      <td>65.0</td>\n",
       "      <td>97.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>15400</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>100.0</td>\n",
       "      <td>96.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>17402</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>300.0</td>\n",
       "      <td>94.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>25023</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>65.0</td>\n",
       "      <td>93.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>25123</td>\n",
       "      <td>Private room</td>\n",
       "      <td>29.0</td>\n",
       "      <td>96.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>995</th>\n",
       "      <td>616278</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>165.0</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>996</th>\n",
       "      <td>617678</td>\n",
       "      <td>Private room</td>\n",
       "      <td>99.0</td>\n",
       "      <td>93.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>997</th>\n",
       "      <td>618082</td>\n",
       "      <td>Private room</td>\n",
       "      <td>45.0</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>998</th>\n",
       "      <td>619010</td>\n",
       "      <td>Private room</td>\n",
       "      <td>20.0</td>\n",
       "      <td>88.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>999</th>\n",
       "      <td>619890</td>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>100.0</td>\n",
       "      <td>90.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "         id        room_type  price  review_scores_rating\n",
       "0     13913     Private room   65.0                  97.0\n",
       "1     15400  Entire home/apt  100.0                  96.0\n",
       "2     17402  Entire home/apt  300.0                  94.0\n",
       "3     25023  Entire home/apt   65.0                  93.0\n",
       "4     25123     Private room   29.0                  96.0\n",
       "..      ...              ...    ...                   ...\n",
       "995  616278  Entire home/apt  165.0                 100.0\n",
       "996  617678     Private room   99.0                  93.0\n",
       "997  618082     Private room   45.0                 100.0\n",
       "998  619010     Private room   20.0                  88.0\n",
       "999  619890  Entire home/apt  100.0                  90.0\n",
       "\n",
       "[1000 rows x 4 columns]"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# querying data from Clickhouse\n",
    "query = \"\"\"\n",
    "SELECT\n",
    "    id,\n",
    "    room_type,\n",
    "    toFloat32OrNull(replaceRegexpAll(price, '[$,]', '')) AS price,\n",
    "    toFloat32OrNull(review_scores_rating) AS review_scores_rating\n",
    "FROM default.ldn_listings\n",
    "WHERE review_scores_rating IS NOT NULL\n",
    "ORDER BY id\n",
    "LIMIT 1000\n",
    "\"\"\"\n",
    "df = ph.read_clickhouse(query, connection=connection)\n",
    "df"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Checking for zero and NaN rating values**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.review_scores_rating.isna().sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df.review_scores_rating.isnull().sum()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Create a scatterplot showing the dependency of an avarage score by a property type**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "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>room_type</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>135.882950</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Hotel room</td>\n",
       "      <td>134.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Private room</td>\n",
       "      <td>54.642437</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Shared room</td>\n",
       "      <td>21.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         room_type       price\n",
       "0  Entire home/apt  135.882950\n",
       "1       Hotel room  134.000000\n",
       "2     Private room   54.642437\n",
       "3      Shared room   21.000000"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# counting mean price\n",
    "mean_price = df.groupby('room_type', as_index=False).price.mean()\n",
    "\n",
    "mean_price"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>room_type</th>\n",
       "      <th>review_scores_rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>93.568787</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Hotel room</td>\n",
       "      <td>92.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Private room</td>\n",
       "      <td>94.766205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Shared room</td>\n",
       "      <td>90.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         room_type  review_scores_rating\n",
       "0  Entire home/apt             93.568787\n",
       "1       Hotel room             92.500000\n",
       "2     Private room             94.766205\n",
       "3      Shared room             90.000000"
      ]
     },
     "execution_count": 13,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# counting mean score\n",
    "mean_score = df.groupby('room_type', as_index=False).review_scores_rating.mean()\n",
    "\n",
    "mean_score"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "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>room_type</th>\n",
       "      <th>price</th>\n",
       "      <th>review_scores_rating</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>135.882950</td>\n",
       "      <td>93.568787</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Hotel room</td>\n",
       "      <td>134.000000</td>\n",
       "      <td>92.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Private room</td>\n",
       "      <td>54.642437</td>\n",
       "      <td>94.766205</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Shared room</td>\n",
       "      <td>21.000000</td>\n",
       "      <td>90.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         room_type       price  review_scores_rating\n",
       "0  Entire home/apt  135.882950             93.568787\n",
       "1       Hotel room  134.000000             92.500000\n",
       "2     Private room   54.642437             94.766205\n",
       "3      Shared room   21.000000             90.000000"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# merging mean score and mean price into one dataframe\n",
    "df_mean = mean_price.merge(mean_score, on='room_type')\n",
    "\n",
    "df_mean"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plotting the data\n",
    "plt.figure(figsize=(12, 8))\n",
    "sns.scatterplot(x=df_mean.price, y=df_mean.review_scores_rating, hue=df_mean.room_type);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Count host verification methods and find the most popular one for advertisements offering experiences**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "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>host_id</th>\n",
       "      <th>host_verifications</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6821</td>\n",
       "      <td>['email', 'phone', 'facebook', 'reviews', 'jum...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11431</td>\n",
       "      <td>['email', 'phone', 'facebook', 'reviews', 'jum...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>27929</td>\n",
       "      <td>['email', 'phone', 'reviews', 'jumio', 'govern...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>54730</td>\n",
       "      <td>['email', 'phone', 'facebook', 'reviews', 'off...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>54987</td>\n",
       "      <td>['email', 'phone', 'google', 'reviews', 'manua...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1356</th>\n",
       "      <td>33365279</td>\n",
       "      <td>['email', 'phone', 'reviews', 'jumio', 'govern...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1357</th>\n",
       "      <td>33585883</td>\n",
       "      <td>['email', 'phone', 'reviews', 'jumio', 'govern...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1358</th>\n",
       "      <td>33657365</td>\n",
       "      <td>['email', 'phone', 'reviews', 'jumio', 'govern...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1359</th>\n",
       "      <td>34219613</td>\n",
       "      <td>['email', 'phone', 'reviews', 'jumio', 'govern...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1360</th>\n",
       "      <td>34299581</td>\n",
       "      <td>['email', 'phone', 'google', 'reviews', 'jumio...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1361 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       host_id                                 host_verifications\n",
       "0         6821  ['email', 'phone', 'facebook', 'reviews', 'jum...\n",
       "1        11431  ['email', 'phone', 'facebook', 'reviews', 'jum...\n",
       "2        27929  ['email', 'phone', 'reviews', 'jumio', 'govern...\n",
       "3        54730  ['email', 'phone', 'facebook', 'reviews', 'off...\n",
       "4        54987  ['email', 'phone', 'google', 'reviews', 'manua...\n",
       "...        ...                                                ...\n",
       "1356  33365279  ['email', 'phone', 'reviews', 'jumio', 'govern...\n",
       "1357  33585883  ['email', 'phone', 'reviews', 'jumio', 'govern...\n",
       "1358  33657365  ['email', 'phone', 'reviews', 'jumio', 'govern...\n",
       "1359  34219613  ['email', 'phone', 'reviews', 'jumio', 'govern...\n",
       "1360  34299581  ['email', 'phone', 'google', 'reviews', 'jumio...\n",
       "\n",
       "[1361 rows x 2 columns]"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# querying data from Clickhouse\n",
    "query = \"\"\"\n",
    "SELECT\n",
    "    DISTINCT(host_id),\n",
    "    host_verifications\n",
    "FROM default.ldn_listings\n",
    "WHERE experiences_offered != 'none'\n",
    "ORDER BY host_id\n",
    "\"\"\"\n",
    "df = ph.read_clickhouse(query, connection=connection)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>host_id</th>\n",
       "      <th>host_verifications</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6821</td>\n",
       "      <td>[email, phone, facebook, reviews, jumio, gover...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>11431</td>\n",
       "      <td>[email, phone, facebook, reviews, jumio, offli...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>27929</td>\n",
       "      <td>[email, phone, reviews, jumio, government_id]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>54730</td>\n",
       "      <td>[email, phone, facebook, reviews, offline_gove...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>54987</td>\n",
       "      <td>[email, phone, google, reviews, manual_offline...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1356</th>\n",
       "      <td>33365279</td>\n",
       "      <td>[email, phone, reviews, jumio, government_id]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1357</th>\n",
       "      <td>33585883</td>\n",
       "      <td>[email, phone, reviews, jumio, government_id]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1358</th>\n",
       "      <td>33657365</td>\n",
       "      <td>[email, phone, reviews, jumio, government_id]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1359</th>\n",
       "      <td>34219613</td>\n",
       "      <td>[email, phone, reviews, jumio, government_id]</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1360</th>\n",
       "      <td>34299581</td>\n",
       "      <td>[email, phone, google, reviews, jumio, offline...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1361 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       host_id                                 host_verifications\n",
       "0         6821  [email, phone, facebook, reviews, jumio, gover...\n",
       "1        11431  [email, phone, facebook, reviews, jumio, offli...\n",
       "2        27929      [email, phone, reviews, jumio, government_id]\n",
       "3        54730  [email, phone, facebook, reviews, offline_gove...\n",
       "4        54987  [email, phone, google, reviews, manual_offline...\n",
       "...        ...                                                ...\n",
       "1356  33365279      [email, phone, reviews, jumio, government_id]\n",
       "1357  33585883      [email, phone, reviews, jumio, government_id]\n",
       "1358  33657365      [email, phone, reviews, jumio, government_id]\n",
       "1359  34219613      [email, phone, reviews, jumio, government_id]\n",
       "1360  34299581  [email, phone, google, reviews, jumio, offline...\n",
       "\n",
       "[1361 rows x 2 columns]"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# changing host verification data type from a string to a list\n",
    "df['host_verifications'] = df.host_verifications.str.strip('[') \\\n",
    "    .str.strip(']') \\\n",
    "    .str.replace(\"'\", '') \\\n",
    "    .str.split(', ')\n",
    "\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>host_id</th>\n",
       "      <th>host_verifications</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6821</td>\n",
       "      <td>email</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6821</td>\n",
       "      <td>phone</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6821</td>\n",
       "      <td>facebook</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6821</td>\n",
       "      <td>reviews</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>6821</td>\n",
       "      <td>jumio</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1360</th>\n",
       "      <td>34299581</td>\n",
       "      <td>jumio</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1360</th>\n",
       "      <td>34299581</td>\n",
       "      <td>offline_government_id</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1360</th>\n",
       "      <td>34299581</td>\n",
       "      <td>selfie</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1360</th>\n",
       "      <td>34299581</td>\n",
       "      <td>government_id</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1360</th>\n",
       "      <td>34299581</td>\n",
       "      <td>identity_manual</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>7976 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "       host_id     host_verifications\n",
       "0         6821                  email\n",
       "0         6821                  phone\n",
       "0         6821               facebook\n",
       "0         6821                reviews\n",
       "0         6821                  jumio\n",
       "...        ...                    ...\n",
       "1360  34299581                  jumio\n",
       "1360  34299581  offline_government_id\n",
       "1360  34299581                 selfie\n",
       "1360  34299581          government_id\n",
       "1360  34299581        identity_manual\n",
       "\n",
       "[7976 rows x 2 columns]"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# exploding host_verification column to count host_verification methods\n",
    "df_unfolded = df.explode('host_verifications')\n",
    "\n",
    "df_unfolded "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "phone                    1361\n",
       "email                    1345\n",
       "reviews                  1309\n",
       "government_id            1196\n",
       "jumio                    1115\n",
       "offline_government_id     469\n",
       "work_email                279\n",
       "facebook                  270\n",
       "selfie                    252\n",
       "identity_manual           227\n",
       "manual_offline             70\n",
       "google                     34\n",
       "kba                        27\n",
       "manual_online              16\n",
       "sent_id                     6\n",
       "Name: host_verifications, dtype: int64"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# counting host verification methods and finding the most popular one\n",
    "df_unfolded.host_verifications.value_counts()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Group data by neighbourhood and experience offered type and count experience offered types**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "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>id</th>\n",
       "      <th>neighbourhood_cleansed</th>\n",
       "      <th>experiences_offered</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>13913</td>\n",
       "      <td>Islington</td>\n",
       "      <td>business</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>15400</td>\n",
       "      <td>Kensington and Chelsea</td>\n",
       "      <td>romantic</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>17506</td>\n",
       "      <td>Hammersmith and Fulham</td>\n",
       "      <td>business</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>38950</td>\n",
       "      <td>Haringey</td>\n",
       "      <td>business</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>41311</td>\n",
       "      <td>Barnet</td>\n",
       "      <td>business</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1543</th>\n",
       "      <td>6550647</td>\n",
       "      <td>Croydon</td>\n",
       "      <td>family</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1544</th>\n",
       "      <td>6551365</td>\n",
       "      <td>Islington</td>\n",
       "      <td>business</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1545</th>\n",
       "      <td>6558281</td>\n",
       "      <td>Camden</td>\n",
       "      <td>business</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1546</th>\n",
       "      <td>6559364</td>\n",
       "      <td>Islington</td>\n",
       "      <td>business</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1547</th>\n",
       "      <td>6574892</td>\n",
       "      <td>Tower Hamlets</td>\n",
       "      <td>social</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1548 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           id  neighbourhood_cleansed experiences_offered\n",
       "0       13913               Islington            business\n",
       "1       15400  Kensington and Chelsea            romantic\n",
       "2       17506  Hammersmith and Fulham            business\n",
       "3       38950                Haringey            business\n",
       "4       41311                  Barnet            business\n",
       "...       ...                     ...                 ...\n",
       "1543  6550647                 Croydon              family\n",
       "1544  6551365               Islington            business\n",
       "1545  6558281                  Camden            business\n",
       "1546  6559364               Islington            business\n",
       "1547  6574892           Tower Hamlets              social\n",
       "\n",
       "[1548 rows x 3 columns]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# querying data from Clickhouse\n",
    "query = \"\"\"\n",
    "SELECT\n",
    "    id,\n",
    "    neighbourhood_cleansed,\n",
    "    experiences_offered\n",
    "FROM default.ldn_listings\n",
    "WHERE experiences_offered != 'none'\n",
    "ORDER BY id\n",
    "\"\"\"\n",
    "df = ph.read_clickhouse(query, connection=connection)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "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>neighbourhood_cleansed</th>\n",
       "      <th>experiences_offered</th>\n",
       "      <th>experiences_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Barking and Dagenham</td>\n",
       "      <td>business</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Barking and Dagenham</td>\n",
       "      <td>social</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Barnet</td>\n",
       "      <td>business</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Barnet</td>\n",
       "      <td>family</td>\n",
       "      <td>7</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Barnet</td>\n",
       "      <td>romantic</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>108</th>\n",
       "      <td>Wandsworth</td>\n",
       "      <td>social</td>\n",
       "      <td>13</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>109</th>\n",
       "      <td>Westminster</td>\n",
       "      <td>business</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>110</th>\n",
       "      <td>Westminster</td>\n",
       "      <td>family</td>\n",
       "      <td>37</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>111</th>\n",
       "      <td>Westminster</td>\n",
       "      <td>romantic</td>\n",
       "      <td>23</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>112</th>\n",
       "      <td>Westminster</td>\n",
       "      <td>social</td>\n",
       "      <td>18</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>113 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "    neighbourhood_cleansed experiences_offered  experiences_count\n",
       "0     Barking and Dagenham            business                  1\n",
       "1     Barking and Dagenham              social                  2\n",
       "2                   Barnet            business                  7\n",
       "3                   Barnet              family                  7\n",
       "4                   Barnet            romantic                  3\n",
       "..                     ...                 ...                ...\n",
       "108             Wandsworth              social                 13\n",
       "109            Westminster            business                 50\n",
       "110            Westminster              family                 37\n",
       "111            Westminster            romantic                 23\n",
       "112            Westminster              social                 18\n",
       "\n",
       "[113 rows x 3 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# counting experience offered\n",
    "df_exp_count = df.groupby(['neighbourhood_cleansed', 'experiences_offered'], as_index=False) \\\n",
    "    .agg({'id': 'count'}) \\\n",
    "    .rename(columns={'id': 'experiences_count'})\n",
    "\n",
    "df_exp_count"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "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>neighbourhood_cleansed</th>\n",
       "      <th>experiences_offered</th>\n",
       "      <th>experiences_count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>36</th>\n",
       "      <td>Hackney</td>\n",
       "      <td>family</td>\n",
       "      <td>61</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>98</th>\n",
       "      <td>Tower Hamlets</td>\n",
       "      <td>business</td>\n",
       "      <td>58</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>109</th>\n",
       "      <td>Westminster</td>\n",
       "      <td>business</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>38</th>\n",
       "      <td>Hackney</td>\n",
       "      <td>social</td>\n",
       "      <td>49</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>39</th>\n",
       "      <td>Hammersmith and Fulham</td>\n",
       "      <td>business</td>\n",
       "      <td>48</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Bromley</td>\n",
       "      <td>family</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>Enfield</td>\n",
       "      <td>social</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Barking and Dagenham</td>\n",
       "      <td>business</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>97</th>\n",
       "      <td>Sutton</td>\n",
       "      <td>romantic</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>83</th>\n",
       "      <td>Newham</td>\n",
       "      <td>romantic</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>100 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "     neighbourhood_cleansed experiences_offered  experiences_count\n",
       "36                  Hackney              family                 61\n",
       "98            Tower Hamlets            business                 58\n",
       "109             Westminster            business                 50\n",
       "38                  Hackney              social                 49\n",
       "39   Hammersmith and Fulham            business                 48\n",
       "..                      ...                 ...                ...\n",
       "12                  Bromley              family                  2\n",
       "30                  Enfield              social                  2\n",
       "0      Barking and Dagenham            business                  1\n",
       "97                   Sutton            romantic                  1\n",
       "83                   Newham            romantic                  1\n",
       "\n",
       "[100 rows x 3 columns]"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# sorting data by experience count and finding the top-100 (most popular ones)\n",
    "df_exp_count_100 = df_exp_count.sort_values('experiences_count', ascending=False) \\\n",
    "    .head(100)\n",
    "df_exp_count_100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th>experiences_offered</th>\n",
       "      <th>business</th>\n",
       "      <th>family</th>\n",
       "      <th>romantic</th>\n",
       "      <th>social</th>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>neighbourhood_cleansed</th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "      <th></th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>Barking and Dagenham</th>\n",
       "      <td>1.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Barnet</th>\n",
       "      <td>7.0</td>\n",
       "      <td>7.0</td>\n",
       "      <td>3.0</td>\n",
       "      <td>5.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Brent</th>\n",
       "      <td>9.0</td>\n",
       "      <td>21.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>6.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Bromley</th>\n",
       "      <td>3.0</td>\n",
       "      <td>2.0</td>\n",
       "      <td>NaN</td>\n",
       "      <td>4.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Camden</th>\n",
       "      <td>36.0</td>\n",
       "      <td>30.0</td>\n",
       "      <td>19.0</td>\n",
       "      <td>25.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "experiences_offered     business  family  romantic  social\n",
       "neighbourhood_cleansed                                    \n",
       "Barking and Dagenham         1.0     NaN       NaN     2.0\n",
       "Barnet                       7.0     7.0       3.0     5.0\n",
       "Brent                        9.0    21.0       5.0     6.0\n",
       "Bromley                      3.0     2.0       NaN     4.0\n",
       "Camden                      36.0    30.0      19.0    25.0"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# pivoting our data by neighbourhood to create a heatmap\n",
    "exp_pivoted = df_exp_count_100.pivot(index='neighbourhood_cleansed',\n",
    "                       columns='experiences_offered',\n",
    "                       values='experiences_count')\n",
    "\n",
    "exp_pivoted.head(5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# creating a heatmap (neighbourhood vs experience_offered type)\n",
    "plt.figure(figsize=(12, 8))\n",
    "sns.heatmap(exp_pivoted, cmap=sns.cubehelix_palette(as_cmap=True));"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Plot price distribution for properties with experieces offered**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>room_type</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Private room</td>\n",
       "      <td>65.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Private room</td>\n",
       "      <td>150.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Private room</td>\n",
       "      <td>45.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Private room</td>\n",
       "      <td>25.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1543</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>100.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1544</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>125.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1545</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>185.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1546</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>110.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1547</th>\n",
       "      <td>Private room</td>\n",
       "      <td>39.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1548 rows × 2 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "            room_type  price\n",
       "0        Private room   65.0\n",
       "1     Entire home/apt  100.0\n",
       "2        Private room  150.0\n",
       "3        Private room   45.0\n",
       "4        Private room   25.0\n",
       "...               ...    ...\n",
       "1543  Entire home/apt  100.0\n",
       "1544  Entire home/apt  125.0\n",
       "1545  Entire home/apt  185.0\n",
       "1546  Entire home/apt  110.0\n",
       "1547     Private room   39.0\n",
       "\n",
       "[1548 rows x 2 columns]"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# querying data from Clickhouse\n",
    "query = \"\"\"\n",
    "SELECT\n",
    "    room_type,\n",
    "    toFloat32OrNull(replaceRegexpAll(price, '[$,]', '')) AS price\n",
    "FROM default.ldn_listings\n",
    "WHERE experiences_offered != 'none'\n",
    "ORDER BY id\n",
    "\"\"\"\n",
    "df = ph.read_clickhouse(query, connection=connection)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 476.5x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plotting data\n",
    "sns.displot(df, x='price', kde=False, hue='room_type');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 476.5x360 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# plotting data in log scale\n",
    "sns.displot(df, x='price', kde=False, hue='room_type', log_scale=True);"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "**Graph an average price variation for properties of different types starting from 2010-01-02**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "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>room_type</th>\n",
       "      <th>price</th>\n",
       "      <th>first_review</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Private room</td>\n",
       "      <td>65.0</td>\n",
       "      <td>2010-08-18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>300.0</td>\n",
       "      <td>2011-03-21</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>65.0</td>\n",
       "      <td>2016-03-05</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Private room</td>\n",
       "      <td>29.0</td>\n",
       "      <td>2010-04-14</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Private room</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2010-10-16</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>995</th>\n",
       "      <td>Private room</td>\n",
       "      <td>45.0</td>\n",
       "      <td>2012-11-07</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>996</th>\n",
       "      <td>Private room</td>\n",
       "      <td>51.0</td>\n",
       "      <td>2013-03-18</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>997</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>79.0</td>\n",
       "      <td>2012-09-17</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>998</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>110.0</td>\n",
       "      <td>2012-09-24</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>999</th>\n",
       "      <td>Private room</td>\n",
       "      <td>70.0</td>\n",
       "      <td>2012-11-26</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 3 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           room_type  price first_review\n",
       "0       Private room   65.0   2010-08-18\n",
       "1    Entire home/apt  300.0   2011-03-21\n",
       "2    Entire home/apt   65.0   2016-03-05\n",
       "3       Private room   29.0   2010-04-14\n",
       "4       Private room  100.0   2010-10-16\n",
       "..               ...    ...          ...\n",
       "995     Private room   45.0   2012-11-07\n",
       "996     Private room   51.0   2013-03-18\n",
       "997  Entire home/apt   79.0   2012-09-17\n",
       "998  Entire home/apt  110.0   2012-09-24\n",
       "999     Private room   70.0   2012-11-26\n",
       "\n",
       "[1000 rows x 3 columns]"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# querying data from Clickhouse\n",
    "query = \"\"\"\n",
    "SELECT\n",
    "    room_type,\n",
    "    toFloat32OrNull(replaceRegexpAll(price, '[$,]', '')) AS price,\n",
    "    first_review\n",
    "FROM default.ldn_listings\n",
    "WHERE first_review > '2010-01-01'\n",
    "LIMIT 1000\n",
    "\"\"\"\n",
    "df = ph.read_clickhouse(query, connection=connection)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>room_type</th>\n",
       "      <th>price</th>\n",
       "      <th>first_review</th>\n",
       "      <th>year</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Private room</td>\n",
       "      <td>65.0</td>\n",
       "      <td>2010-08-18</td>\n",
       "      <td>2010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>300.0</td>\n",
       "      <td>2011-03-21</td>\n",
       "      <td>2011</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>65.0</td>\n",
       "      <td>2016-03-05</td>\n",
       "      <td>2016</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Private room</td>\n",
       "      <td>29.0</td>\n",
       "      <td>2010-04-14</td>\n",
       "      <td>2010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Private room</td>\n",
       "      <td>100.0</td>\n",
       "      <td>2010-10-16</td>\n",
       "      <td>2010</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>995</th>\n",
       "      <td>Private room</td>\n",
       "      <td>45.0</td>\n",
       "      <td>2012-11-07</td>\n",
       "      <td>2012</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>996</th>\n",
       "      <td>Private room</td>\n",
       "      <td>51.0</td>\n",
       "      <td>2013-03-18</td>\n",
       "      <td>2013</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>997</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>79.0</td>\n",
       "      <td>2012-09-17</td>\n",
       "      <td>2012</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>998</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>110.0</td>\n",
       "      <td>2012-09-24</td>\n",
       "      <td>2012</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>999</th>\n",
       "      <td>Private room</td>\n",
       "      <td>70.0</td>\n",
       "      <td>2012-11-26</td>\n",
       "      <td>2012</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>1000 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "           room_type  price first_review  year\n",
       "0       Private room   65.0   2010-08-18  2010\n",
       "1    Entire home/apt  300.0   2011-03-21  2011\n",
       "2    Entire home/apt   65.0   2016-03-05  2016\n",
       "3       Private room   29.0   2010-04-14  2010\n",
       "4       Private room  100.0   2010-10-16  2010\n",
       "..               ...    ...          ...   ...\n",
       "995     Private room   45.0   2012-11-07  2012\n",
       "996     Private room   51.0   2013-03-18  2013\n",
       "997  Entire home/apt   79.0   2012-09-17  2012\n",
       "998  Entire home/apt  110.0   2012-09-24  2012\n",
       "999     Private room   70.0   2012-11-26  2012\n",
       "\n",
       "[1000 rows x 4 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 'resampling' to years\n",
    "df = df.assign(year = df.first_review.apply(lambda x: x[0:4]))\n",
    "\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "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>room_type</th>\n",
       "      <th>year</th>\n",
       "      <th>price</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>2010</td>\n",
       "      <td>103.500000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>Private room</td>\n",
       "      <td>2010</td>\n",
       "      <td>56.428570</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>Hotel room</td>\n",
       "      <td>2010</td>\n",
       "      <td>88.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>2011</td>\n",
       "      <td>128.792450</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>Private room</td>\n",
       "      <td>2011</td>\n",
       "      <td>52.607143</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>2012</td>\n",
       "      <td>135.467255</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>Private room</td>\n",
       "      <td>2012</td>\n",
       "      <td>53.269840</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>24</th>\n",
       "      <td>Shared room</td>\n",
       "      <td>2012</td>\n",
       "      <td>22.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>2013</td>\n",
       "      <td>137.887634</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>Private room</td>\n",
       "      <td>2013</td>\n",
       "      <td>57.036587</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>17</th>\n",
       "      <td>Private room</td>\n",
       "      <td>2014</td>\n",
       "      <td>56.842106</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>25</th>\n",
       "      <td>Shared room</td>\n",
       "      <td>2014</td>\n",
       "      <td>20.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>2014</td>\n",
       "      <td>154.583328</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>2015</td>\n",
       "      <td>159.380951</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>18</th>\n",
       "      <td>Private room</td>\n",
       "      <td>2015</td>\n",
       "      <td>48.058823</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>2016</td>\n",
       "      <td>99.733330</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>19</th>\n",
       "      <td>Private room</td>\n",
       "      <td>2016</td>\n",
       "      <td>57.307693</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>2017</td>\n",
       "      <td>111.666664</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20</th>\n",
       "      <td>Private room</td>\n",
       "      <td>2017</td>\n",
       "      <td>51.142857</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>Hotel room</td>\n",
       "      <td>2017</td>\n",
       "      <td>180.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>2018</td>\n",
       "      <td>125.066666</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>21</th>\n",
       "      <td>Private room</td>\n",
       "      <td>2018</td>\n",
       "      <td>55.416668</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>2019</td>\n",
       "      <td>301.666656</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>22</th>\n",
       "      <td>Private room</td>\n",
       "      <td>2019</td>\n",
       "      <td>74.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>23</th>\n",
       "      <td>Private room</td>\n",
       "      <td>2020</td>\n",
       "      <td>39.000000</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Entire home/apt</td>\n",
       "      <td>2020</td>\n",
       "      <td>50.000000</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "          room_type  year       price\n",
       "0   Entire home/apt  2010  103.500000\n",
       "13     Private room  2010   56.428570\n",
       "11       Hotel room  2010   88.000000\n",
       "1   Entire home/apt  2011  128.792450\n",
       "14     Private room  2011   52.607143\n",
       "2   Entire home/apt  2012  135.467255\n",
       "15     Private room  2012   53.269840\n",
       "24      Shared room  2012   22.000000\n",
       "3   Entire home/apt  2013  137.887634\n",
       "16     Private room  2013   57.036587\n",
       "17     Private room  2014   56.842106\n",
       "25      Shared room  2014   20.000000\n",
       "4   Entire home/apt  2014  154.583328\n",
       "5   Entire home/apt  2015  159.380951\n",
       "18     Private room  2015   48.058823\n",
       "6   Entire home/apt  2016   99.733330\n",
       "19     Private room  2016   57.307693\n",
       "7   Entire home/apt  2017  111.666664\n",
       "20     Private room  2017   51.142857\n",
       "12       Hotel room  2017  180.000000\n",
       "8   Entire home/apt  2018  125.066666\n",
       "21     Private room  2018   55.416668\n",
       "9   Entire home/apt  2019  301.666656\n",
       "22     Private room  2019   74.000000\n",
       "23     Private room  2020   39.000000\n",
       "10  Entire home/apt  2020   50.000000"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# calculating an average price\n",
    "df_avg = df.groupby(['room_type', 'year'], as_index=False) \\\n",
    "    .agg({'price': 'mean'}) \\\n",
    "    .sort_values('year')\n",
    "\n",
    "df_avg"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x576 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "# graphing data\n",
    "plt.figure(figsize=(12, 8))\n",
    "sns.lineplot(data=df_avg, x='year', y='price', hue='room_type');"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# LESSON 7 -- Dashboards"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create a table for our retail data\n",
    "CREATE TABLE test.retail (\n",
    "    InvoiceNo stringMergeTree,\n",
    "    StockCode stringMergeTree,\n",
    "    Description stringMergeTree,\n",
    "    Quantity Int32,\n",
    "    InvoiceDate DateTime('Europe/London'),\n",
    "    UnitPrice Decimal64(3),\n",
    "    CustomerID uint32,\n",
    "    Country stringMergeTree)\n",
    "ENGINE = MergeTree\n",
    "ORDER BY InvoiceDate, CustomerID"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Check the number of unique customers by country\n",
    "SELECT COUNT(DISTINCT CustomerID) AS uniq_customers,\n",
    "       Country\n",
    "FROM default.retail\n",
    "GROUP BY Country"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Check the change of MAU (monthly active users) in UK, Australia and Netherlands\n",
    "SELECT COUNT(DISTINCT CustomerID) AS active_users,\n",
    "       toStartOfMonth(InvoiceDate) AS MONTH,\n",
    "       Country AS country\n",
    "FROM default.retail\n",
    "WHERE country IN ('United Kingdom',\n",
    "                  'Australia',\n",
    "                  'Netherlands')\n",
    "GROUP BY MONTH,\n",
    "         country\n",
    "ORDER BY MONTH"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Check the change of MAU (monthly active users) for all countries except UK\n",
    "SELECT COUNT(DISTINCT CustomerID) AS active_users,\n",
    "       toStartOfMonth(InvoiceDate) AS MONTH,\n",
    "       Country AS country\n",
    "FROM default.retail\n",
    "WHERE country != 'United Kingdom'\n",
    "GROUP BY MONTH,\n",
    "         country\n",
    "ORDER BY MONTH"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate AOV (average order value) for each country\n",
    "SELECT AVG(order_value) AS AOV,\n",
    "       Country\n",
    "FROM\n",
    "  (SELECT InvoiceNo,\n",
    "          SUM(TotalPrice) AS order_value,\n",
    "          Country\n",
    "   FROM\n",
    "     (SELECT InvoiceNo,\n",
    "             Quantity * UnitPrice AS TotalPrice,\n",
    "             Country\n",
    "      FROM default.retail)\n",
    "   GROUP BY InvoiceNo,\n",
    "            Country\n",
    "   ORDER BY order_value DESC)\n",
    "GROUP BY Country\n",
    "ORDER BY AOV DESC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Check the dynamics of AOV metrics for United Kingdom, Germany, France, Spain,\n",
    "# Netherlands, Belgium, Switzerland, Portugal, Australia, USA\n",
    "SELECT AVG(order_value) AS AOV,\n",
    "       Country,\n",
    "       MONTH\n",
    "FROM\n",
    "  (SELECT SUM(TotalPrice) AS order_value,\n",
    "          MONTH,\n",
    "          Country\n",
    "   FROM\n",
    "     (SELECT InvoiceNo,\n",
    "             Quantity * UnitPrice AS TotalPrice,\n",
    "             Country,\n",
    "             toStartOfMonth(InvoiceDate) AS MONTH\n",
    "      FROM default.retail\n",
    "      WHERE Country IN ('United Kingdom',\n",
    "                        'Germany',\n",
    "                        'France',\n",
    "                        'Spain',\n",
    "                        'Netherlands',\n",
    "                        'Belgium',\n",
    "                        'Switzerland',\n",
    "                        'Portugal',\n",
    "                        'Australia',\n",
    "                        'USA') )\n",
    "   GROUP BY InvoiceNo,\n",
    "            Country,\n",
    "            MONTH\n",
    "   ORDER BY MONTH ASC, order_value DESC)\n",
    "GROUP BY MONTH,\n",
    "         Country\n",
    "ORDER BY MONTH ASC, AOV DESC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate an average number of items per order by country\n",
    "SELECT AVG(quantity_per_invoice) AS average_items,\n",
    "       Country\n",
    "FROM\n",
    "  (SELECT InvoiceNo,\n",
    "          SUM(Quantity) AS quantity_per_invoice,\n",
    "          Country\n",
    "   FROM default.retail\n",
    "   GROUP BY InvoiceNo,\n",
    "            Country)\n",
    "GROUP BY Country\n",
    "ORDER BY average_items DESC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Investigate customers from the Netherlands. \n",
    "# Find a customer who has bought the biggest number of items\n",
    "SELECT\n",
    "    SUM(Quantity) AS overall_quantity,\n",
    "    CustomerID\n",
    "FROM default.retail\n",
    "WHERE Country == 'Netherlands'\n",
    "GROUP BY CustomerID\n",
    "ORDER BY overall_quantity DESC\n",
    "LIMIT 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate revenue deistribution per country\n",
    "SELECT\n",
    "    Country,\n",
    "    --UnitPrice, \n",
    "    --Quantity,\n",
    "    SUM(UnitPrice * Quantity) AS total_revenue\n",
    "FROM default.retail\n",
    "GROUP BY Country\n",
    "ORDER BY total_revenue DESC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate monthly revenue dynamics for UK\n",
    "SELECT\n",
    "    toStartOfMonth(InvoiceDate) AS month,\n",
    "    SUM(UnitPrice * Quantity) AS monthly_reveue\n",
    "FROM default.retail\n",
    "WHERE Country == 'United Kingdom'\n",
    "    AND month < '2011-12-01'\n",
    "GROUP BY month\n",
    "ORDER BY month ASC\n",
    "LIMIT 100"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Find the top-10 sales value items in UK in November\n",
    "SELECT\n",
    "    Description,\n",
    "    SUM(Quantity) AS items_sold,\n",
    "    UnitPrice,\n",
    "    items_sold * UnitPrice AS total_sales_revenue\n",
    "FROM default.retail\n",
    "WHERE (toStartOfMonth(InvoiceDate) == '2011-11-01')\n",
    "    AND Country == 'United Kingdom'\n",
    "GROUP BY Description, UnitPrice\n",
    "ORDER BY total_sales_revenue DESC\n",
    "LIMIT 10"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# LESSON 8 -- Window Functions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# PostgreSQL via Redash was used for making queries."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Find accumulative volume of organic avocado sold in New York and Los Angeles. \n",
    "# Sort the results by date in ascedning order.\n",
    "SELECT \n",
    "  region, \n",
    "  date, \n",
    "  total_volume, \n",
    "  SUM(total_volume) OVER w AS volume \n",
    "FROM \n",
    "  avocado \n",
    "WHERE \n",
    "  region IN ('NewYork', 'LosAngeles') \n",
    "  AND type = 'organic' \n",
    "GROUP BY \n",
    "  region, \n",
    "  date, \n",
    "  total_volume WINDOW w AS (\n",
    "    PARTITION BY region \n",
    "    ORDER BY \n",
    "      date ASC\n",
    "  ) \n",
    "ORDER BY \n",
    "  region DESC, \n",
    "  date ASC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Find accumulative volume of organic avocado sold in New York and Los Angeles year by year. \n",
    "# Sort the results by date in ascedning order.\n",
    "SELECT \n",
    "  region, \n",
    "  date, \n",
    "  year, \n",
    "  total_volume, \n",
    "  SUM(total_volume) OVER w AS volume \n",
    "FROM \n",
    "  avocado \n",
    "WHERE \n",
    "  region IN ('NewYork', 'LosAngeles') \n",
    "  AND type = 'organic' \n",
    "GROUP BY \n",
    "  region, \n",
    "  date, \n",
    "  year, \n",
    "  total_volume WINDOW w AS (\n",
    "    PARTITION BY region, \n",
    "    year \n",
    "    ORDER BY \n",
    "      date ASC\n",
    "  ) \n",
    "ORDER BY \n",
    "  region DESC, \n",
    "  date ASC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Find the weekly difference in total volume for conventional avocado sold in US.\n",
    "# Sort the results by date in ascedning order.\n",
    "SELECT \n",
    "  date, \n",
    "  total_volume, \n",
    "  region, \n",
    "  type, \n",
    "  total_volume - lag(total_volume, 1) OVER w AS week_diff \n",
    "FROM \n",
    "  avocado \n",
    "WHERE \n",
    "  region = 'TotalUS' \n",
    "  AND type = 'conventional' \n",
    "GROUP BY \n",
    "  date, \n",
    "  region, \n",
    "  type, \n",
    "  total_volume WINDOW w AS (\n",
    "    ORDER BY \n",
    "      date ASC\n",
    "  ) \n",
    "ORDER BY \n",
    "  date ASC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Find the weekly difference in total volume avocado sold in New York in 2018. \n",
    "# Comapre the results for organic and conventional avocado types.\n",
    "SELECT \n",
    "  date, \n",
    "  total_volume, \n",
    "  type, \n",
    "  total_volume - lag(total_volume, 1) OVER w AS week_diff \n",
    "FROM \n",
    "  avocado \n",
    "WHERE \n",
    "  region = 'NewYork' \n",
    "  AND year = 2018 \n",
    "GROUP BY \n",
    "  date, \n",
    "  type, \n",
    "  total_volume WINDOW w AS (\n",
    "    PARTITION BY type \n",
    "    ORDER BY \n",
    "      date ASC\n",
    "  ) \n",
    "ORDER BY \n",
    "  type ASC, \n",
    "  date ASC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Calculate the rolling average price (window=2) for conventional avocado in New York.\n",
    "SELECT \n",
    "  date, \n",
    "  average_price, \n",
    "  region, \n",
    "  type, \n",
    "  AVG(average_price) OVER w AS rolling_price \n",
    "FROM \n",
    "  avocado \n",
    "WHERE \n",
    "  region = 'NewYork' \n",
    "  AND type = 'conventional' \n",
    "GROUP BY \n",
    "  date, \n",
    "  region, \n",
    "  type, \n",
    "  average_price WINDOW w AS (\n",
    "    ORDER BY \n",
    "      date ASC ROWS BETWEEN 2 PRECEDING \n",
    "      AND 0 FOLLOWING\n",
    "  ) \n",
    "ORDER BY \n",
    "  type ASC, \n",
    "  date ASC"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# LESSON 9"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "tbc.."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}