How to check for NaN values

To check for NaN (Not a Number) values in Python, especially when working with data, there are several ways depending on the libraries you're using. NaN typically indicates missing or undefined data, and catching it early is key to maintaining data integrity.

Here are some common and effective ways to check for NaN values:

 Using the math module

import math

x = float('nan')

print(math.isnan(x))  # True

• math.isnan() is useful for scalar (single) values.
  
• Works well for general numeric computations where no external libraries are used.
  

 Using numpy

import numpy as np

arr = np.array([1, 2, np.nan])

print(np.isnan(arr))  # [False False  True]

• np.isnan() is vectorized and ideal for arrays or lists of numbers.
  
• Returns a Boolean array indicating which values are NaN.
  

 Using pandas

import pandas as pd

data = pd.Series([1, 2, None, float('nan')])

print(data.isna())

• isna() (or isnull()) is perfect for detecting NaNs in Series/DataFrames.
  
• Widely used in data analysis and cleaning workflows.
  

 Key Notes:

• None and NaN are not always the same — pandas treats them similarly, but basic Python does not.
  
• Always clean or handle NaN values before performing calculations or visualizations.
  

In short, use math.isnan() for simple checks, numpy.isnan() for numerical arrays, and pandas.isna() for structured data. These tools make NaN detection both simple and powerful.