3 Easy Ways to Create a Subset of Python Dataframe

Python Pandas module provides us with two data structures, namely, Series and Dataframe to store the values.

A Dataframe is a data structure that holds the data in the form of a matrix i.e. it contains the data in the value-form of rows and columns. Thus, in association with it, we can create and access the subset of it in the below formats:

• Access data according to the rows as subset
• Fetch data according to the columns as subset
• Access specific data from some rows as well as columns as subset

Having understood about Dataframe and subsets, let us now understand the different techniques to create a subset out of a Dataframe.

To create subsets of a dataframe, we need to create a dataframe. Let’s get that out of our way first:

import pandas as pd 
data = {"Roll-num": [10,20,30,40,50,60,70], "Age":[12,14,13,12,14,13,15], "NAME":['John','Camili','Rheana','Joseph','Amanti','Alexa','Siri']}
block = pd.DataFrame(data)
print("Original Data frame:\n")
print(block)

Output:

Original Data frame:

   Roll-num  Age    NAME
0        10   12    John
1        20   14  Camili
2        30   13  Rheana
3        40   12  Joseph
4        50   14  Amanti
5        60   13   Alexa
6        70   15    Siri

Here, we have created a data frame using pandas.DataFrame() method. We will be using the above created dataset throughout this article

Let us begin!

1. Create a subset of a Python dataframe using the loc() function

Python loc() function enables us to form a subset of a data frame according to a specific row or column or a combination of both.

The loc() function works on the basis of labels i.e. we need to provide it with the label of the row/column to choose and create the customized subset.

Syntax:

pandas.dataframe.loc[]

Example 1: Extract data of specific rows of a dataframe

block.loc[[0,1,3]]

Output:

As seen below, we have created a subset which includes all the data of row 0, 1, and 3.

Roll-num	Age	NAME
0	10	12	John
1	20	14	Camili
3	40	12	Joseph

Example 2: Create a subset of rows using slicing

block.loc[0:3]

Here, we have extracted the data of all the rows from index 0 to index 3 using slicing operator with loc() function.

Output:

Roll-num	Age	NAME
0	10	12	John
1	20	14	Camili
2	30	13	Rheana
3	40	12	Joseph

Example 3: Create a subset of particular columns using labels

block.loc[0:2,['Age','NAME']]

Output:

Age	NAME
0	12	John
1	14	Camili
2	13	Rheana

Here, we have created a subset which includes data from rows 0 to 2, but includes that of only some specific columns i.e. ‘Age’ and ‘NAME’.

2. Using Python iloc() function to create a subset of a dataframe

Python iloc() function enables us to create subset choosing specific values from rows and columns based on indexes.

That is, unlike loc() function which works on labels, iloc() function works on index values. We can choose and create a subset of a Python dataframe from the data providing the index numbers of the rows and columns.

Syntax:

pandas.dataframe.iloc[]

Example:

block.iloc[[0,1,3,6],[0,2]]

Here, we have created a subset which includes the data of the rows 0,1,3 and 6 as well as column number 0 and 2 i.e. ‘Roll-num’ and ‘NAME’.

Output:

Roll-num	NAME
0	10	John
1	20	Camili
3	40	Joseph
6	70	Siri

3. Indexing operator to create a subset of a dataframe

In a simple manner, we can make use of an indexing operator i.e. square brackets to create a subset of the data.

Syntax:

dataframe[['col1','col2','colN']]

Example:

block[['Age','NAME']]

Here, we have selected all the data values of the columns ‘Age’ and ‘NAME’, respectively.

Output:

Age	NAME
0	12	John
1	14	Camili
2	13	Rheana
3	12	Joseph
4	14	Amanti
5	13	Alexa
6	15	Siri

By this, we have come to the end of this topic. Feel free to comment below, in case you come across any question. For more such posts related to Python, stay tuned, and till then, Happy Learning!! :)