Skip to content

As a Data Engineer at an electronics e-commerce company, Voltmart, you have been requested by a peer Machine Learning team to clean the data containing the information about orders made last year. They are planning to further use this cleaned data to build a demand forecasting model. To achieve this, they have shared their requirements regarding the desired output table format.

An analyst shared a parquet file called "orders_data.parquet" for you to clean and preprocess.

You can see the dataset schema below along with the cleaning requirements:

orders_data.parquet

columndata typedescriptioncleaning requirements
order_datetimestampDate and time when the order was madeModify: Remove orders placed between 12am and 5am (inclusive); convert from timestamp to date
time_of_daystringPeriod of the day when the order was madeNew column containing (lower bound inclusive, upper bound exclusive): "morning" for orders placed 5-12am, "afternoon" for orders placed 12-6pm, and "evening" for 6-12pm
order_idlongOrder IDN/A
productstringName of a product orderedRemove rows containing "TV" as the company has stopped selling this product; ensure all values are lowercase
product_eandoubleProduct IDN/A
categorystringBroader category of a productEnsure all values are lowercase
purchase_addressstringAddress line where the order was made ("House Street, City, State Zipcode")N/A
purchase_statestringUS State of the purchase addressNew column containing: the State that the purchase was ordered from
quantity_orderedlongNumber of product units orderedN/A
price_eachdoublePrice of a product unitN/A
cost_pricedoubleCost of production per product unitN/A
turnoverdoubleTotal amount paid for a product (quantity x price)N/A
margindoubleProfit made by selling a product (turnover - cost)N/A

from pyspark.sql import (
    SparkSession,
    types,
    functions as F,
)

spark = (
    SparkSession
    .builder
    .appName('cleaning_orders_dataset_with_pyspark')
    .getOrCreate()
)
orders_data = spark.read.parquet('orders_data.parquet')
orders_data.toPandas().head()
from pyspark.sql.functions import *
orders_data = orders_data.filter(F.hour(orders_data.order_date) > 5)

orders_data = orders_data.withColumn("time_of_day",
                      when(F.hour("order_date") < 12, "morning")
                      .when(F.hour("order_date") < 18, "afternoon")
                      .otherwise("evening")
                      )
orders_data = orders_data.withColumn("order_date", to_date("order_date"))
orders_data = orders_data.withColumn("product", F.lower("product"))
orders_data = orders_data.filter(~orders_data.product.contains("tv"))
orders_data = orders_data.withColumn("category", F.lower("category"))
orders_data = orders_data.withColumn("purchase_address_split", F.split("purchase_address", ","))
orders_data = orders_data.withColumn("purchase_state_with_zip", orders_data.purchase_address_split.getItem(F.size("purchase_address_split")-1))
orders_data = orders_data.drop("purchase_address_split")
from pyspark.sql.functions import udf
from pyspark.sql.types import StringType

# Define the UDF
def get_first_two_items(s):
    return s[1:3]

# Register the UDF
get_first_two_items_udf = udf(get_first_two_items, StringType())

# Apply the UDF to a specific column, for example 'product'
orders_data_clean = orders_data.withColumn('purchase_state', get_first_two_items_udf(orders_data['purchase_state_with_zip']))
orders_data_clean.toPandas().head()
unique_purchase_states = orders_data_clean.select("purchase_state").distinct()
unique_purchase_states.toPandas().head()
‌
‌
‌