melt() for unpivoting in Pandas - Practice Problems & Coding Challenges

Challenge - 5 Problems

❓ Predict Output

intermediate

Output of melt() with id_vars and value_vars

What is the output DataFrame after running the following code?

Pandas

import pandas as pd

df = pd.DataFrame({
    'City': ['NY', 'LA'],
    '2019': [100, 200],
    '2020': [110, 210]
})

result = pd.melt(df, id_vars=['City'], value_vars=['2019', '2020'], var_name='Year', value_name='Population')
print(result)

A

  City  Year  Population
0   NY  2019         100
1   LA  2019         200
2   NY  2020         110
3   LA  2020         210

B

  City  Year  Population
0   NY  2019         110
1   LA  2019         210
2   NY  2020         100
3   LA  2020         200

C

  City  Year  Population
0   NY  2019         100
1   NY  2020         110
2   LA  2019         200
3   LA  2020         210

D

  City  Year  Population
0   NY  2019         200
1   LA  2019         100
2   NY  2020         210
3   LA  2020         110

💡 Hint

Remember melt keeps id_vars as is and stacks value_vars into rows.

The melt function keeps 'City' as identifier and unpivots '2019' and '2020' columns into 'Year' and 'Population'. The order is by rows then columns.

❓ data_output

intermediate

Number of rows after melting

Given a DataFrame with 3 rows and 4 columns, if you melt it using 1 column as id_vars and 3 columns as value_vars, how many rows will the melted DataFrame have?

Pandas

import pandas as pd

df = pd.DataFrame({
    'ID': [1, 2, 3],
    'A': [10, 20, 30],
    'B': [40, 50, 60],
    'C': [70, 80, 90]
})

melted = pd.melt(df, id_vars=['ID'], value_vars=['A', 'B', 'C'])
print(len(melted))

A 12

B 3

C 9

D 6

💡 Hint

Each row is repeated for each value_var column.

There are 3 rows and 3 value_vars, so total rows after melt = 3 * 3 = 9.

🔧 Debug

advanced

Identify the error in melt usage

What error will this code raise and why?

Pandas

import pandas as pd

df = pd.DataFrame({
    'Name': ['Alice', 'Bob'],
    'Math': [90, 80],
    'Science': [85, 95]
})

result = pd.melt(df, id_vars='Name', value_vars='Math,Science')
print(result)

A No error, prints melted DataFrame

B KeyError: 'Math,Science'

C ValueError: value_vars must be list-like

D TypeError: id_vars must be list-like, not str

💡 Hint

Check the type of value_vars argument.

value_vars='Math,Science' is a single string, so pandas looks for a column named exactly 'Math,Science' which does not exist, causing KeyError.

🚀 Application

advanced

Using melt to reshape sales data

You have a DataFrame with columns: 'Store', 'Jan_Sales', 'Feb_Sales', 'Mar_Sales'. You want to reshape it so that each row shows 'Store', 'Month', and 'Sales'. Which melt call achieves this?

Pandas

import pandas as pd

df = pd.DataFrame({
    'Store': ['A', 'B'],
    'Jan_Sales': [100, 150],
    'Feb_Sales': [110, 160],
    'Mar_Sales': [120, 170]
})

A pd.melt(df, id_vars=['Store'], var_name='Month', value_name='Sales')

B pd.melt(df, id_vars=['Store'], value_vars=['Store'], var_name='Month', value_name='Sales')

C pd.melt(df, value_vars=['Jan_Sales', 'Feb_Sales', 'Mar_Sales'], var_name='Month', value_name='Sales')

D pd.melt(df, id_vars=['Store'], value_vars=['Jan_Sales', 'Feb_Sales', 'Mar_Sales'], var_name='Month', value_name='Sales')

💡 Hint

Specify id_vars and value_vars explicitly to keep 'Store' and unpivot sales columns.

Option D correctly keeps 'Store' as id_vars and melts the sales columns into 'Month' and 'Sales'. Option D melts all columns including 'Store', causing wrong output.

🧠 Conceptual

expert

Understanding melt with multiple id_vars

Given a DataFrame with columns ['Region', 'Year', 'Q1', 'Q2'], what does melt(df, id_vars=['Region', 'Year'], value_vars=['Q1', 'Q2']) do?

A Creates a DataFrame with columns ['Region', 'Year', 'variable', 'value'], where 'Q1' and 'Q2' values are stacked under 'value' for each Region and Year.

B Creates a DataFrame with columns ['Region', 'Year', 'Q1', 'Q2'] unchanged.

C Creates a DataFrame with columns ['Region', 'variable', 'value'] dropping 'Year'.

D Creates a DataFrame with columns ['Year', 'variable', 'value'] dropping 'Region'.

💡 Hint

melt keeps all id_vars columns as is and stacks value_vars into two columns.

melt keeps 'Region' and 'Year' columns as identifiers and unpivots 'Q1' and 'Q2' into 'variable' and 'value' columns.