Dataframe.melt() in pandas

In this tutorial, we are going to learn how to use the melt() method in Pandas. The demonstration is done by using various examples. This is used to change the shape of the existing data frame. This is done for the process of data analysis. Here, the data frame appears in a long format rather than a wide format. So let’s begin the tutorial.

Parameters of dataframe.melt() in Pandas

id_vars: Columns that should be used in the final result.

value_vars: Columns used to unpivot. If no value is specified, all the columns of  id_vars that are not set are considered.

var_name: The name that can be specified for the variable column.

value_name: The name that can be used to change the name of the value column.

col_level: This is used in the case of multi-level indexed columns.

Data Frame

The below data frame is created to demonstrate the melt() method.

import pandas as p
data1 = { 'Age':[20,5,89,6],
          'Gender':['M','M','F','M'],
          'Qualified':['y','n','y','n'],
          'Height':[5,3,6,3.2]
        }
d1 = p.DataFrame(data1) 
print(d1)

OUTPUT:

  Age Gender Qualified Height
0 20  M      y         5.0
1 5   M      n         3.0
2 89  F      y         6.0
3 6   M      n         3.2

Using melt() without any parameters

When melt() method is used without any parameters, all columns are melted to long form.

import pandas as p
data1 = { 'Age':[20,5,89,6],
          'Gender':['M','M','F','M'],
          'Qualified':['y','n','y','n'],
          'Height':[5,3,6,3.2]
        }
d1 = p.DataFrame(data1) 
print(p.melt(d1))

OUTPUT:

  variable   value
0 Age        20
1 Age        5
2 Age        89
3 Age        6
4 Gender     M
5 Gender     M
6 Gender     F
7 Gender     M
8 Qualified  y
9 Qualified  n
10 Qualified y
11 Qualified n
12 Height    5
13 Height    3
14 Height    6
15 Height    3.2

Using id_vars parameter

Here the columns Age and Height are used as id_vars columns.

import pandas as p
data1 = { 'Age':[20,5,89,6],
          'Gender':['M','M','F','M'],
          'Qualified':['y','n','y','n'],
          'Height':[5,3,6,3.2]
        }
d1 = p.DataFrame(data1) 
print(p.melt(d1, id_vars=['Age','Height']))

OUTPUT:

 Age Height variable  value
0 20 5.0    Gender    M
1 5  3.0    Gender    M
2 89 6.0    Gender    F
3 6  3.2    Gender    M
4 20 5.0    Qualified y
5 5  3.0    Qualified n
6 89 6.0    Qualified y
7 6  3.2    Qualified n

Using value_vars and var_name

Here, the values are displayed for the height column and the name of the column is changed as name.

import pandas as p
data1 = { 'Age':[20,5,89,6],
          'Gender':['M','M','F','M'],
          'Qualified':['y','n','y','n'],
          'Height':[5,3,6,3.2]
        }
d1 = p.DataFrame(data1) 
print(p.melt(d1, id_vars=['Age'],value_vars=['Height'],var_name='Name'))

OUTPUT:

  Age Name   value
0 20  Height 5.0
1 5   Height 3.0
2 89  Height 6.0
3 6   Height 3.2

Using value_name

Here, we change the name of the values column.

import pandas as p
data1 = { 'Age':[20,5,89,6],
          'Gender':['M','M','F','M'],
          'Qualified':['y','n','y','n'],
          'Height':[5,3,6,3.2]
        }
d1 = p.DataFrame(data1) 
print(p.melt(d1, id_vars=['Age'],value_vars=['Height'],var_name='Name',value_name='Height1'))

OUTPUT:

  Age Name   Height1
0 20  Height 5.0
1 5   Height 3.0
2 89  Height 6.0
3 6   Height 3.2

Also read: Dataframe.describe() in Pandas