Skip to content

pandas-use-of-dot-is-null (PD003)

Derived from the pandas-vet linter.

What it does

Checks for uses of .isnull on Pandas objects.

Why is this bad?

In the Pandas API, .isna and .isnull are equivalent. For consistency, prefer .isna over .isnull.

As a name, .isna more accurately reflects the behavior of the method, since these methods check for NaN and NaT values in addition to None values.

Example

import pandas as pd

animals_df = pd.read_csv("animals.csv")
pd.isnull(animals_df)

Use instead:

import pandas as pd

animals_df = pd.read_csv("animals.csv")
pd.isna(animals_df)

References