melt() for unpivoting in Pandas - Step-by-Step Execution

Concept Flow - melt() for unpivoting

Start with wide DataFrame

↓

Choose id_vars (fixed columns)

↓

Choose value_vars (columns to unpivot)

↓

Apply melt()

↓

Get long DataFrame with variable and value columns

↓

Use for easier analysis or plotting

The melt() function takes a wide table and turns selected columns into rows, making data longer and easier to analyze.

Execution Sample

Pandas

import pandas as pd

df = pd.DataFrame({
    'Name': ['Anna', 'Bob'],
    'Math': [90, 80],
    'Science': [85, 88]
})

melted = pd.melt(df, id_vars=['Name'], value_vars=['Math', 'Science'], var_name='Subject', value_name='Score')

This code changes a table of student scores from wide to long format, listing each subject and score per student.

Execution Table

Step	DataFrame State	Action	Resulting DataFrame Preview
1	Original df	Create DataFrame with columns Name, Math, Science	[Name, Math, Science] [Anna, 90, 85] [Bob, 80, 88]
2	df	Call pd.melt with id_vars=['Name'], value_vars=['Math', 'Science']	New columns: Name, Subject, Score
3	melted	Rows created for each subject per student	[Name, Subject, Score] [Anna, Math, 90] [Anna, Science, 85] [Bob, Math, 80] [Bob, Science, 88]
4	melted	Data is now long format, easier for analysis	Ready for plotting or grouping

💡 All specified value_vars columns are unpivoted into rows, process complete.

Variable Tracker

Variable	Start	After melt()
df	[Anna, 90, 85] [Bob, 80, 88]	Unchanged
melted	Not defined	[Anna, Math, 90] [Anna, Science, 85] [Bob, Math, 80] [Bob, Science, 88]

Key Moments - 2 Insights

Why do we specify id_vars and value_vars in melt()?

id_vars are columns to keep fixed (like 'Name'), value_vars are columns to turn into rows (like 'Math' and 'Science'). See execution_table step 2 where melt() is called with these parameters.

What happens to the original columns after melt()?

Columns in value_vars are removed from wide form and appear as values in the new 'variable' column (here named 'Subject'). See execution_table step 3 for the new DataFrame structure.

Visual Quiz - 3 Questions

Look at the execution table, what is the value of 'Subject' for the second row in the melted DataFrame?

A Name

B Science

C Math

D Score

💡 Hint

Check execution_table row 3 showing the melted DataFrame preview.

At which step does the DataFrame change from wide to long format?

A Step 1

B Step 2

C Step 3

D Step 4

💡 Hint

Look at execution_table rows 2 and 3 to see when the DataFrame structure changes.

If we add another subject column 'English' to value_vars, how would the melted DataFrame change?

A More rows per student, one for English scores

B Fewer rows, English scores replace Science

C No change, English ignored

D Columns 'Name' and 'English' swapped

💡 Hint

Recall that melt() creates one row per value_vars column per id_vars row, see variable_tracker for melted shape.

Concept Snapshot

pandas melt() unpivots data:
- id_vars: columns to keep fixed
- value_vars: columns to turn into rows
- var_name: name for new variable column
- value_name: name for new value column
Transforms wide data to long format for easier analysis.

Full Transcript

The melt() function in pandas changes a wide table into a long one by turning selected columns into rows. We start with a DataFrame with columns like Name, Math, and Science. We keep 'Name' fixed using id_vars, and unpivot 'Math' and 'Science' using value_vars. The result is a longer DataFrame with columns Name, Subject, and Score. Each student has one row per subject. This format is easier for grouping and plotting. Key points are choosing which columns to keep and which to unpivot. The execution table shows each step from original data to melted data. The variable tracker shows how the DataFrame changes. The quiz checks understanding of the new structure and how melt works.