Divide a Pandas DataFrame randomly in a given ratio

Divide a Pandas Dataframe task is very useful in case of split a given dataset into train and test data for training and testing purposes in the field of Machine Learning, Artificial Intelligence, etc.  Let’s see how to divide the pandas dataframe randomly into given ratios. For this task, We will use Dataframe.sample() and Dataframe.drop() methods of pandas dataframe together.

The Syntax of these functions are as follows – 

• Dataframe.sample()

Syntax: DataFrame.sample(n=None, frac=None, replace=False, weights=None, random_state=None, axis=None)

Return Type: A new object of same type as caller containing n items randomly sampled from the caller object.

• Dataframe.drop()

Syntax: DataFrame.drop(labels=None, axis=0, index=None, columns=None, level=None, inplace=False, errors=’raise’)

Return: Dataframe with dropped values.

Example: Now, let’s create a Dataframe:

Python3

# Importing required libraries
import pandas as pd

record = {
    'course_name': ['Data Structures', 'Python',
                    'Machine Learning', 'Web Development'],
    'student_name': ['Ankit', 'Shivangi',
                     'Priya', 'Shaurya'],
    'student_city': ['Chennai', 'Pune',
                     'Delhi', 'Mumbai'],
    'student_gender': ['M', 'F',
                       'F', 'M'] }

# Creating a dataframe
df = pd.DataFrame(record)

# show the dataframe
df

Output: 

Dataframe

Example 1: Divide a Dataframe randomly into a 1:1 ratio.

Python3

# Importing required libraries
import pandas as pd

record = {
    'course_name': ['Data Structures', 'Python',
                    'Machine Learning', 'Web Development'],
    'student_name': ['Ankit', 'Shivangi',
                     'Priya', 'Shaurya'],
    'student_city': ['Chennai', 'Pune',
                     'Delhi', 'Mumbai'],
    'student_gender': ['M', 'F',
                       'F', 'M'] }

# Creating a dataframe
df = pd.DataFrame(record)

# Creating a dataframe with 50%
# values of original dataframe
part_50 = df.sample(frac = 0.5)

# Creating dataframe with
# rest of the 50% values
rest_part_50 = df.drop(part_50.index)

print("\n50% of the given DataFrame:")
print(part_50)

print("\nrest 50% of the given DataFrame:")
print(rest_part_50)

Output: 

Divide dataframe

Example 2: Divide a Dataframe randomly into a 3:1 ratio.

Python3

# Importing required libraries
import pandas as pd

record = {
    'course_name': ['Data Structures', 'Python',
                    'Machine Learning', 'Web Development'],
    'student_name': ['Ankit', 'Shivangi',
                     'Priya', 'Shaurya'],
    'student_city': ['Chennai', 'Pune',
                     'Delhi', 'Mumbai'],
    'student_gender': ['M', 'F',
                       'F', 'M'] }

# Creating a dataframe
df = pd.DataFrame(record)

# Creating a dataframe with 75%
# values of original dataframe
part_75 = df.sample(frac = 0.75)

# Creating dataframe with
# rest of the 25% values
rest_part_25 = df.drop(part_75.index)

print("\n75% of the given DataFrame:")
print(part_75)

print("\nrest 25% of the given DataFrame:")
print(rest_part_25)

Output: 

Divide Dataframe