Replace NaN values in a Column in Pandas

This tutorial will discuss about unique ways to replace nan values in a column in pandas.

Table Of Contents

Preparing DataSet

Create a DataFrame with certain rows and columns.

Let’s see the complete example,

import pandas as pd
import numpy as np

# List of Tuples
employees= [('Mark',   'US',    'Tech',   5),
            ('Riti',    np.NaN, 'Tech' ,  7),
            ('Shanky',  np.NaN, 'PMO' ,   2),
            ('Shreya',  'UK',   'Design', 2),
            ('Aadi',    np.NaN, 'Tech',   11),
            ('Sim',     'US',   'Tech',   4)]

# Create a DataFrame object from list of tuples
df = pd.DataFrame(employees,
                  columns=['Name', 'Location', 'Team', 'Experience'])

print(df)

Output

     Name Location    Team  Experience
0    Mark       US    Tech           5
1    Riti      NaN    Tech           7
2  Shanky      NaN     PMO           2
3  Shreya       UK  Design           2
4    Aadi      NaN    Tech          11
5     Sim       US    Tech           4

Replace NaN values using fillna()

Select the column, and call fillna() method on it. It accepts a replacement value as first argument, and replaces all the NaN values in the column with that replacement value. Also pass the inplace parameter as True, so that it changes the column in place. Let’s see an example,

Here we will replace all the NaN values in column Location with default value Australia.

# Replace all NaN values  in column 'Location'
# with string 'Australia'
df['Location'].fillna('Australia', inplace=True)

print(df)

Output:

     Name   Location    Team  Experience
0    Mark         US    Tech           5
1    Riti  Australia    Tech           7
2  Shanky  Australia     PMO           2
3  Shreya         UK  Design           2
4    Aadi  Australia    Tech          11
5     Sim         US    Tech           4

Replace NaN values replace fillna()

Select the column, and call replace() method on it. Pass the value to be replaced i.e. np.NaN, and replacement string i.e. Australia as parameters in it. Also, pass the inplace parameter as True in it, so that all modifications are done in place in the column.

Here it will replace all the NaN values in column Location with default value Australia.

# Replace all NaN values  in column 'Location'
# with string 'Australia'
df['Location'].replace(np.nan, 'Australia', inplace=True)

print(df)

Output:

     Name   Location    Team  Experience
0    Mark         US    Tech           5
1    Riti  Australia    Tech           7
2  Shanky  Australia     PMO           2
3  Shreya         UK  Design           2
4    Aadi  Australia    Tech          11
5     Sim         US    Tech           4

Summary

We learned about different ways to replace NaN values in DataFrame column with given value.