Group by and aggregate

pandaseasygroupby

Write revenue_by_category(df) returning a Series indexed by category with the total revenue (units * price), sorted descending.

preloaded fixtures

import pandas as pd
import numpy as np

sales = pd.DataFrame({
    "order_id": [1, 2, 3, 4, 5, 6, 7, 8],
    "customer": ["ada", "grace", "ada", "alan", "grace", "marie", "alan", "ada"],
    "country": ["UK", "USA", "UK", "UK", "USA", "France", "UK", "UK"],
    "category": ["hardware", "audio", "audio", "furniture", "hardware", "audio", "hardware", "furniture"],
    "units": [1, 2, 1, 3, 1, 2, 4, 1],
    "price": [129.0, 249.99, 199.0, 420.0, 329.5, 249.99, 59.0, 540.0],
    "rating": [4.5, np.nan, 3.0, 5.0, np.nan, 4.0, 4.5, 2.5],
})

customers = pd.DataFrame({
    "customer": ["ada", "grace", "alan", "hedy"],
    "segment": ["pro", "pro", "starter", "starter"],
})

hint ladder

sql ↔ pandas

solution.py

loading editor…

step inspector

Inspect a chained expression (df.query(…).groupby(…).agg(…)) to see the shape and preview after each step.

output

Run your code to see its output, or check your solution to grade it.

Reference solution

The shape we check against. Any implementation that passes the assertions is valid — this one favours clarity.

def revenue_by_category(df):
    rev = df['units'] * df['price']
    return rev.groupby(df['category']).sum().sort_values(ascending=False)

Code is blurred until you solve this problem — the reasoning stays readable.