Grouping by multiple columns in Pandas - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of grouping by two columns and summing

What is the output of this code snippet that groups data by two columns and sums the values?

Pandas

import pandas as pd

data = pd.DataFrame({
    'City': ['NY', 'LA', 'NY', 'LA', 'NY'],
    'Category': ['A', 'A', 'B', 'B', 'A'],
    'Sales': [100, 200, 150, 300, 50]
})
result = data.groupby(['City', 'Category']).sum()
print(result)

A

              Sales
City Category       
LA   A          300
     B          300
NY   A          150
     B          150

B

              Sales
City Category       
LA   A          200
     B          300
NY   A          150
     B          150

C

              Sales
City Category       
LA   A          200
     B          300
NY   A          100
     B          150

D

              Sales
City Category       
LA   A          200
     B          450
NY   A          150
     B          150

💡 Hint

Think about how groupby sums values for each unique pair of City and Category.

The groupby sums Sales for each City and Category pair. For LA and B, sales is 300 only, not 450. For NY and A, sales is 100 + 50 = 150.

❓ data_output

intermediate

Number of groups formed by grouping on two columns

Given this DataFrame, how many groups will be formed when grouping by 'Type' and 'Color'?

Pandas

import pandas as pd

data = pd.DataFrame({
    'Type': ['Fruit', 'Fruit', 'Vegetable', 'Fruit', 'Vegetable'],
    'Color': ['Red', 'Green', 'Green', 'Red', 'Red'],
    'Quantity': [10, 15, 7, 5, 3]
})
groups = data.groupby(['Type', 'Color'])
print(len(groups))

A 3

B 6

C 5

D 4

💡 Hint

Count unique pairs of Type and Color in the data.

The unique pairs are: (Fruit, Red), (Fruit, Green), (Vegetable, Green), (Vegetable, Red). So 4 groups.

🔧 Debug

advanced

Identify the error in grouping by multiple columns

What error will this code raise when trying to group by multiple columns?

Pandas

import pandas as pd

data = pd.DataFrame({
    'A': [1, 2, 1],
    'B': [3, 4, 3],
    'C': [5, 6, 7]
})
result = data.groupby('A', 'B').sum()
print(result)

A ValueError: No axis named 'B'

B KeyError: 'B'

C AttributeError: 'DataFrame' object has no attribute 'groupby'

D No error, outputs grouped sum

💡 Hint

Check how groupby expects multiple columns as input.

groupby expects a list or array-like for multiple columns, not multiple positional arguments.

❓ visualization

advanced

Visualizing grouped data by multiple columns

Which option correctly creates a bar plot showing the sum of 'Value' grouped by 'Group' and 'Subgroup'?

Pandas

import pandas as pd
import matplotlib.pyplot as plt

data = pd.DataFrame({
    'Group': ['X', 'X', 'Y', 'Y', 'X'],
    'Subgroup': ['A', 'B', 'A', 'B', 'A'],
    'Value': [10, 20, 30, 40, 50]
})
grouped = data.groupby(['Group', 'Subgroup']).sum()

# Which code below plots the grouped data correctly?

A

grouped.plot(kind='bar')
plt.show()

B

grouped.plot(kind='bar', stacked=True)
plt.show()

C

grouped.unstack().plot(kind='bar')
plt.show()

D

data.plot(kind='bar', x='Group', y='Value')
plt.show()

💡 Hint

Think about how to reshape grouped data for a clear bar plot with multiple categories.

unstack() reshapes the grouped data so that Subgroup becomes columns, allowing a grouped bar plot.

🚀 Application

expert

Calculate weighted average after grouping by multiple columns

You have this DataFrame with 'Category', 'Subcategory', 'Score', and 'Weight'. How do you calculate the weighted average Score for each Category and Subcategory group?

Pandas

import pandas as pd

data = pd.DataFrame({
    'Category': ['A', 'A', 'B', 'B', 'A'],
    'Subcategory': ['X', 'X', 'Y', 'Y', 'Z'],
    'Score': [80, 90, 70, 60, 85],
    'Weight': [1, 2, 1, 3, 2]
})

A data.groupby(['Category', 'Subcategory']).apply(lambda x: (x['Score'] * x['Weight']).sum() / x['Weight'].sum())

B data.groupby(['Category', 'Subcategory']).mean('Score')

C data.groupby(['Category', 'Subcategory']).agg({'Score': 'mean', 'Weight': 'sum'})

D data.groupby(['Category', 'Subcategory']).sum()['Score'] / data.groupby(['Category', 'Subcategory']).sum()['Weight']

💡 Hint

Weighted average is sum of (value * weight) divided by sum of weights.

Option A correctly applies weighted average calculation per group using apply and lambda.