Testing for NaN in Python

In this tutorial, we will explore how to test for ‘Not a Number’ (NaN) values in Python. NaN values usually arise from undefined or unrepresentable data like 0/0.

They can cause a variety of problems if not properly handled when performing calculations or data analysis.

Python provides several methods for identifying these NaN values within a dataset, so let’s get started!

Step 1: Import Necessary Libraries

The primary library we will use for this tutorial is the pandas library. This library provides the isnull function, which returns True for each NaN or None value in a given series or dataframe.

1
import pandas as pd

Step 2: Create a DataFrame with NaN Values

Let’s create an exemplar dataframe with NaN values to work on. We can create NaN values using the numpy library.

1
2
3
4
import numpy as np
 
data = {'A': [1,2,np.nan], 'B': [5,np.nan,1], 'C': [1,2,3]}
df = pd.DataFrame(data)

Step 3: Use the pandas isnull Function

Now we will use the isnull function, which returns a Boolean value for each element in the DataFrame or Series. If the element is NaN or None, isnull returns True; otherwise, it returns False.

1
print(df.isnull())

Full Code

1
2
3
4
5
6
7
8
9
import pandas as pd
import numpy as np
 
# Create a DataFrame
data = {'A': [1,2,np.nan], 'B': [5,np.nan,1], 'C': [1,2,3]}
df = pd.DataFrame(data)
 
# Use isnull function to check for NaN values
print(df.isnull())

Output

       A      B      C
0  False  False  False
1  False   True  False
2   True  False  False

From the output above, the True values represent NaN or None values found in the DataFrame.

Conclusion

This tutorial has shown you how to test for NaN values in Python using the pandas isnull function. Remember, always handling these kinds of values in your datasets is crucial as they can cause unexpected results during your data analysis or when developing machine learning models.

A proper understanding and handling of NaN values are necessary for developing more accurate models.