How to Use melt in pandas: Simple Guide with Examples

Use pandas.melt() to transform a DataFrame from wide format to long format by unpivoting selected columns into rows. Specify id_vars to keep columns fixed and value_vars to select columns to unpivot. This helps in data analysis and visualization by making data tidy.

Syntax

The pandas.melt() function has this basic syntax:

• frame: The DataFrame to reshape.
• id_vars: Columns to keep fixed (not unpivoted).
• value_vars: Columns to unpivot into rows (optional, defaults to all columns not in id_vars).
• var_name: Name for the new variable column (optional).
• value_name: Name for the new value column (optional).

python

pandas.melt(frame, id_vars=None, value_vars=None, var_name=None, value_name='value', col_level=None, ignore_index=True)

Example

This example shows how to convert a wide DataFrame with sales data for different years into a long format using melt(). It keeps the Product column fixed and unpivots the year columns.

python

import pandas as pd

data = {
    'Product': ['A', 'B', 'C'],
    '2019': [100, 150, 200],
    '2020': [110, 160, 210],
    '2021': [120, 170, 220]
}
df = pd.DataFrame(data)

melted = pd.melt(df, id_vars=['Product'], var_name='Year', value_name='Sales')
print(melted)

Output

  Product  Year  Sales
0       A  2019    100
1       B  2019    150
2       C  2019    200
3       A  2020    110
4       B  2020    160
5       C  2020    210
6       A  2021    120
7       B  2021    170
8       C  2021    220

Common Pitfalls

Common mistakes when using melt() include:

• Not specifying id_vars, which can cause unexpected columns to be unpivoted.
• Forgetting to set var_name or value_name, leading to unclear column names.
• Trying to melt columns with mixed data types without cleaning first.

Here is an example of a wrong and right way:

python

import pandas as pd

data = {'Name': ['John', 'Anna'], 'Math': [90, 95], 'Science': [85, 88]}
df = pd.DataFrame(data)

# Wrong: Not specifying id_vars (Name will be unpivoted too)
wrong = pd.melt(df)

# Right: Specify id_vars to keep 'Name' fixed
right = pd.melt(df, id_vars=['Name'], var_name='Subject', value_name='Score')

print('Wrong melt output:\n', wrong)
print('\nRight melt output:\n', right)

Output

Wrong melt output:
    variable  value
0      Name   John
1      Name   Anna
2      Math     90
3      Math     95
4   Science     85
5   Science     88

Right melt output:
    Name  Subject  Score
0   John     Math     90
1   Anna     Math     95
2   John  Science     85
3   Anna  Science     88

Quick Reference

Remember these tips when using melt():

• Use id_vars to keep columns fixed.
• Use value_vars to select columns to unpivot.
• Set var_name and value_name for clear column names.
• Check data types before melting to avoid errors.

Key Takeaways

Use pandas.melt() to reshape data from wide to long format by unpivoting columns.

Always specify id_vars to keep certain columns fixed during melting.

Set var_name and value_name for meaningful column names in the output.

Check your data types before melting to avoid unexpected results.

Melted data is easier to analyze and visualize in many cases.