apply() on rows (axis=1) in Pandas - Cheat Sheet & Quick Revision

Recall & Review

beginner

What does the apply() function do when used with axis=1 in pandas?

It applies a function to each row of the DataFrame, allowing you to perform row-wise operations.

beginner

How do you specify that apply() should work on rows instead of columns?

By setting the parameter axis=1 inside the apply() function.

beginner

Example: What will this code do?<br>

df.apply(lambda row: row['A'] + row['B'], axis=1)

It will add the values in columns 'A' and 'B' for each row and return a Series with these sums.

intermediate

Why might you use apply() with axis=1 instead of vectorized operations?

To perform custom or complex operations that involve multiple columns in each row, which are not easily done with simple vectorized functions.

beginner

What type of object does the function inside apply(axis=1) receive as input?

It receives a pandas Series representing a single row of the DataFrame.

What does axis=1 mean when used with apply() in pandas?

A Apply function only to the first row

B Apply function to each column

C Apply function to each row

D Apply function to the entire DataFrame at once

Setting axis=1 tells pandas to apply the function to each row.

What type of object is passed to the function inside apply(axis=1)?

A A Series representing a row

B A DataFrame

C A list

D A scalar value

The function receives a Series object representing one row of the DataFrame.

Which of these is a correct way to sum columns 'A' and 'B' for each row using apply()?

A df.apply(lambda x: x['A'] + x['B'], axis=1)

B df.apply(lambda x: x['A'] + x['B'], axis=0)

C df.apply(sum, axis=1)

D df.sum(axis=1)

Using apply() with axis=1 and a lambda function sums columns 'A' and 'B' row-wise.

Why might apply(axis=1) be slower than vectorized operations?

A Because it does not support lambda functions

B Because it processes columns instead of rows

C Because it only works on small DataFrames

D Because it uses loops internally

apply(axis=1) often uses Python-level loops, which are slower than vectorized operations.

If you want to create a new column based on row-wise calculations, which method is suitable?

A df['new'] = df.apply(func, axis=0)

B df['new'] = df.apply(func, axis=1)

C df['new'] = df.sum(axis=0)

D df['new'] = df.sum(axis=1)

Using apply() with axis=1 lets you compute values row-wise for a new column.

Explain how apply() with axis=1 works in pandas and give a simple example.

Think about how you would add two columns for each row.

Describe a situation where using apply() with axis=1 is better than vectorized operations.

Consider when you need custom calculations per row.