transform() for group-level operations in Pandas - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of group transform with mean subtraction

What is the output of the following code snippet?

Pandas

import pandas as pd

df = pd.DataFrame({'Team': ['A', 'A', 'B', 'B'], 'Points': [10, 20, 30, 40]})
df['AdjPoints'] = df.groupby('Team')['Points'].transform(lambda x: x - x.mean())
print(df)

A

  Team  Points  AdjPoints
0    A      10       -5.0
1    A      20        5.0
2    B      30       -5.0
3    B      40        5.0

B

  Team  Points  AdjPoints
0    A      10       10.0
1    A      20       20.0
2    B      30       30.0
3    B      40       40.0

C

  Team  Points  AdjPoints
0    A      10       15.0
1    A      20       15.0
2    B      30       35.0
3    B      40       35.0

D

  Team  Points  AdjPoints
0    A      10       -10.0
1    A      20       10.0
2    B      30       -30.0
3    B      40       30.0

💡 Hint

Think about what subtracting the group mean from each value does.

The transform applies the lambda function to each group. For each group, it subtracts the mean of that group's Points from each Points value, centering the data around zero within each group.

❓ data_output

intermediate

Number of items after transform with duplicated values

Given the code below, how many rows does the resulting DataFrame have?

Pandas

import pandas as pd

df = pd.DataFrame({'Category': ['X', 'X', 'Y', 'Y', 'Y'], 'Value': [1, 2, 3, 4, 5]})
df['Rank'] = df.groupby('Category')['Value'].transform('rank')
print(len(df))

A 5

B 3

C 2

D 1

💡 Hint

transform keeps the original DataFrame shape.

The transform method returns a Series with the same length as the original DataFrame, so the number of rows remains unchanged.

🔧 Debug

advanced

Identify the error in transform usage

What error does the following code raise?

Pandas

import pandas as pd

df = pd.DataFrame({'Group': ['G1', 'G1', 'G2'], 'Score': [10, 20, 30]})
df['Scaled'] = df.groupby('Group')['Score'].transform(lambda x: x / x.sum())
print(df)

A TypeError: unsupported operand type(s) for /: 'int' and 'method'

B No error, outputs scaled scores per group

C AttributeError: 'SeriesGroupBy' object has no attribute 'sum'

D ValueError: Length of values does not match length of index

💡 Hint

Check if the lambda function uses correct pandas methods.

The code correctly divides each score by the sum of scores in its group. The transform returns a Series aligned with the original DataFrame, so no error occurs.

🚀 Application

advanced

Using transform to fill missing values with group mean

You have a DataFrame with missing values in the 'Sales' column. Which code snippet correctly fills missing 'Sales' values with the mean sales of their group in 'Region'?

Pandas

import pandas as pd
import numpy as np

df = pd.DataFrame({'Region': ['East', 'East', 'West', 'West'], 'Sales': [100, np.nan, 200, np.nan]})

A df['Sales'] = df['Sales'].fillna(df.groupby('Region')['Sales'].mean())

B df['Sales'] = df.groupby('Region')['Sales'].transform('mean')

C df['Sales'] = df.groupby('Region')['Sales'].apply(lambda x: x.fillna(x.mean()))

D df['Sales'] = df.groupby('Region')['Sales'].transform(lambda x: x.fillna(x.mean()))

💡 Hint

transform keeps the original index and shape, apply may change it.

Option D uses transform with a lambda that fills missing values with the group mean, preserving the DataFrame shape and index alignment. Option D tries to fillna with a Series that does not align properly. Option D uses apply which may change the index. Option D replaces all values with the group mean, not just missing ones.

🧠 Conceptual

expert

Why use transform instead of apply for group-level operations?

Which statement best explains why transform is preferred over apply for group-level operations that return a Series with the same shape as the original DataFrame?

A apply always returns a DataFrame, transform always returns a scalar.

B apply is faster than transform but cannot handle group operations.

C transform returns a Series aligned with the original DataFrame, preserving index and shape, while apply may return aggregated or differently shaped results.

D transform can only be used with numeric data, apply works with all data types.

💡 Hint

Think about the shape and alignment of the output.

transform returns a Series with the same length as the original DataFrame, aligned by index, which is useful for adding group-level calculations back to the original data. apply can return aggregated results or different shapes, making it less suitable for this use case.