Pandas GroupBy Aggregation Examples

import pandas as pd

df = pd.read_csv("orders.csv")

# Basic groupby with single aggregation
revenue_by_region = df.groupby("region")["amount"].sum()

# Multiple aggregations
stats = df.groupby("region")["amount"].agg(["sum", "mean", "count", "std"])

# Named aggregations (clean output)
summary = df.groupby("region").agg(
    total_revenue=pd.NamedAgg(column="amount", aggfunc="sum"),
    avg_order=pd.NamedAgg(column="amount", aggfunc="mean"),
    order_count=pd.NamedAgg(column="id", aggfunc="count"),
    unique_customers=pd.NamedAgg(column="customer_id", aggfunc="nunique"),
).reset_index()

# Group by multiple columns
monthly = df.groupby(["region", pd.Grouper(key="date", freq="ME")])["amount"].sum()

# Transform: add group-level calculation as a new column
df["pct_of_region"] = df.groupby("region")["amount"].transform(
    lambda x: x / x.sum() * 100
)

# Filter groups: keep only regions with total > 10000
big_regions = df.groupby("region").filter(lambda g: g["amount"].sum() > 10000)

# Rank within groups
df["rank_in_region"] = df.groupby("region")["amount"].rank(
    ascending=False, method="dense"
)

# Pivot table
pivot = pd.pivot_table(
    df, values="amount", index="region", columns="status",
    aggfunc="sum", fill_value=0, margins=True
)

print(summary)
print(pivot)