7.2 Multiple Keys and Named Agg

By Ravindra Bagale Last updated: 30 September 2026

summary = (
    df.groupby(["city", "festival"], as_index=False)
      .agg(
          gmv=("amount", "sum"),
          orders=("order_id", "nunique"),
          avg_ticket=("amount", "mean"),
      )
)
summary.sort_values("gmv", ascending=False)

Steps in Jupyter

Create df as above.
Run groupby("city")["amount"].sum().
Build the named aggregation summary.
Filter summary[summary["festival"] == "Diwali"].

What you should see. Pune Diwali and Nashik Diwali rows with higher GMV than “None” days — fictional spike pattern.

Ravindra Bagale's Tip

Khup students groupby nantar sum() karun Series milavatat aani column name haravtat. Report sathi as_index=False aani named agg (gmv=("amount","sum")) vapra — columns clear rahtat. Lakshat theva!