Level 3: Aggregating · Exercise 34 of 63 · Pro

Share of the group total with transform

A pandas exercise from Pandasdrill. Solve it with a Python editor in your browser running real pandas, and it is checked the moment you submit.

The task

Add a `share` column with each order's revenue as a fraction of its region's total revenue, and return the columns `order_id`, `region`, `revenue` and `share` for every order, in the original order and with the original labels.

What this exercise teaches

transform keeps the shape of the input, so each order gets its own region's total next to it and the division lines up row by row. A plain .sum() collapses to one row per region, and dividing by that aligns on the index labels, which are regions on one side and order positions on the other, so you get NaN everywhere.

A hint if you get stuck

groupby(...)[col].transform("sum") returns the group total repeated on every row of that group, aligned with the original index.