Applying Functions

What is .apply()?

• .apply() applies a function to a Pandas object (Series or DataFrame).
• It works elementwise, rowwise, or columnwise, depending on how you use it.
• It’s more flexible than vectorised operations, but often slower.

On a Series

When used on a Series, the function is applied elementwise:

import pandas as pd​s = pd.Series([1, 2, 3, 4])​s.apply(lambda x: x**2)

Output:

0     11     42     93    16dtype: int64

On a DataFrame

When used on a DataFrame, you can choose axis:

• axis=0 (default) → apply function to each column.
• axis=1 → apply function to each row.

Column example:

df = pd.DataFrame({    "A": [1, 2, 3],    "B": [10, 20, 30]})​# Apply columnwise (axis=0)df.apply(sum, axis=0)

Output:

A     6B    60dtype: int64

Row example:

# Apply rowwise (axis=1)df.apply(lambda row: row["B"] - row["A"], axis=1)

Output:

0     91    182    27dtype: int64

Returning New Columns

You can assign results back to new columns:

df["Diff"] = df.apply(lambda row: row["B"] - row["A"], axis=1)

Returning DataFrames

If your function returns a Series, .apply() can expand it into multiple columns:

def stats(row):    return pd.Series({"Sum": row.sum(), "Mean": row.mean()})​df_stats = df.apply(stats, axis=1)