pandas-tasks-01-read-order_items.py

import pandas as pd

# Reading order_items

order_items_path = "/Users/itversity/Research/data/retail_db/order_items/part-00000"
order_items_schema = [
    "order_item_id",
    "order_item_order_id",
    "order_item_product_id",
    "order_item_quantity",
    "order_item_subtotal",
    "order_item_product_price"
]

order_items = pd.read_csv(
    order_items_path,
    header=None,
    names=order_items_schema
)

pandas-tasks-02-projection.py

# Project order_item_order_id and order_item_subtotal

order_items[["order_item_order_id", "order_item_subtotal"]]

pandas-tasks-03-filtering.py

# Filter for order_item_order_id 2
order_items[order_items.order_item_order_id == 2]
order_items[order_items["order_item_order_id"] == 2]

pandas-tasks-04-aggregation.py

# Compute revenue using order_item_subtotal for each order_item_order_id
order_items.groupby("order_item_order_id")["order_item_subtotal"]. \
  sum()

# Compute revenue using order_item_subtotal for each order_item_order_id with alias
order_items.groupby("order_item_order_id")["order_item_subtotal"]. \
  agg(order_revenue="sum")

pandas-tasks-05-joins.py

# Reading orders

orders_path = "/Users/itversity/Research/data/retail_db/orders/part-00000"
orders_schema = [
    "order_id",
    "order_date",
    "order_customer_id",
    "order_status"
]

orders = pd.read_csv(
    orders_path,
    header=None,
    names=orders_schema
)

# Join orders and order_items

orders.set_index("order_id"). \
  join(order_items.set_index("order_item_order_id"))

pandas-tasks-06-compute-daily-revenue.py

# Compute Daily Revenue using 
# orders.order_date and order_items.order_item_order_subtotal 
# considering only COMPLETE and CLOSED orders.

orders_filtered = orders[orders.order_status.isin(["COMPLETE", "CLOSED"])]
orders_join = orders_filtered.set_index("order_id"). \
  join(order_items.set_index("order_item_order_id"))
daily_revenue = orders_join.groupby("order_date")["order_item_subtotal"].agg(revenue="sum").round(2)
daily_revenue