Pandas: Unpivot Wide DataFrames with melt()

Unpivot wide-format DataFrames into long format using the Pandas pd.melt() function. Transform columns into rows, specify identifier variables, customize output column names, and prepare clean datasets for analysis.

Show Table of Contents ↓

Introduction to pd.melt() 🔝

The pd.melt() function unpivots a DataFrame from a wide format (where attributes are spread across multiple columns) into a long format (where attributes are gathered into rows). It is widely used in data cleaning and preparing inputs for machine learning models or plotting libraries.

Create wide-format sample data 🔝

Let us start with a wide-format student dataset containing names, IDs, and subject marks across separate columns.

import pandas as pd
df = pd.DataFrame({
    "NAME": ["Ravi", "Raju", "Alex"],
    "ID": [1, 2, 3],
    "MATH": [30, 40, 50],
    "ENGLISH": [20, 30, 40]
})
print(df)
assert df.shape == (3, 4)

Expected output

   NAME  ID  MATH  ENGLISH
0  Ravi   1    30       20
1  Raju   2    40       30
2  Alex   3    50       40

Basic melting with id_vars 🔝

Use id_vars to keep specific columns as identifier variables while all other remaining columns are unpivoted into rows.

melted = pd.melt(df, id_vars=["NAME", "ID"])
print(melted)
assert len(melted) == 6

Expected output

   NAME  ID variable  value
0  Ravi   1     MATH     30
1  Raju   2     MATH     40
2  Alex   3     MATH     50
3  Ravi   1  ENGLISH     20
4  Raju   2  ENGLISH     30
5  Alex   3  ENGLISH     40

Selecting specific value columns 🔝

By default, all columns not specified in id_vars are melted. You can explicitly restrict which columns to unpivot using value_vars.

specific = pd.melt(df, id_vars=["NAME"], value_vars=["MATH"])
print(specific)
assert len(specific) == 3

Expected output

   NAME variable  value
0  Ravi     MATH     30
1  Raju     MATH     40
2  Alex     MATH     50

Customizing variable and value names 🔝

Rename the default variable and value output columns using the var_name and value_name parameters.

custom = pd.melt(df, id_vars=["NAME"], value_vars=["MATH", "ENGLISH"], 
                  var_name="SUBJECT", value_name="MARKS")
print(custom)
assert "SUBJECT" in custom.columns and "MARKS" in custom.columns

Expected output

   NAME  SUBJECT  MARKS
0  Ravi     MATH     30
1  Raju     MATH     40
2  Alex     MATH     50
3  Ravi  ENGLISH     20
4  Raju  ENGLISH     30
5  Alex  ENGLISH     40

Resetting index during melt 🔝

By default, original row index labels are preserved in the melted output. Set ignore_index=True to generate a clean RangeIndex.

reset = pd.melt(df, id_vars=["NAME"], value_vars=["MATH"], ignore_index=True)
print(reset)
assert reset.index.tolist() == [0, 1, 2]

Expected output

   NAME variable  value
0  Ravi     MATH     30
1  Raju     MATH     40
2  Alex     MATH     50

Exercise: melt student marks 🔝

Melt the student DataFrame using NAME and ID as identifier variables, renaming the variable column to COURSE and the value column to SCORE, with ignore_index=True.

Exercise solution 🔝

Solution applying all parameters together.

answer = pd.melt(df, id_vars=["NAME", "ID"], value_vars=["MATH", "ENGLISH"],
                  var_name="COURSE", value_name="SCORE", ignore_index=True)
print(answer)
assert len(answer) == 6 and "COURSE" in answer.columns

Expected output

   NAME  ID   COURSE  SCORE
0  Ravi   1     MATH     30
1  Raju   2     MATH     40
2  Alex   3     MATH     50
3  Ravi  ENGLISH     20
4  Raju  ENGLISH     30
5  Alex  ENGLISH     40

Practice in Google Colab 🔝

Open in Google Colab View on GitHub

Run examples in order, inspect melted structures and try the exercise. Save a copy to keep edits.

Reference: Pandas melt() documentation.

Subhendu Mohapatra

Author & Instructor at plus2net

I write and maintain practical tutorials on Python, PHP, SQL, JavaScript, HTML, jQuery, and web development at plus2net. The tutorials focus on clear explanations, working examples, and code that readers can test and adapt while learning.

About the author Watch on YouTube

Connect: