How to insert an isna() clause in a Pandas query

Question

I would like to create an isna() clause using the .query() method in Pandas

I am getting an error though.

For a reproducible example:

import pandas as pd
import seaborn as sns

mpg = sns.load_dataset('mpg')

mpg[mpg['cylinders'].isna()] # This works

mpg.query('cylinders.isna()') # This raises an exception
TypeError: 'Series' objects are mutable, thus they cannot be hashed

Does this answer your question? [Querying for NaN and other names in Pandas](https://stackoverflow.com/questions/26535563/querying-for-nan-and-other-names-in-pandas) — Leonid Mednikov, Sep 21 '22 at 11:22

score 7 · Accepted Answer · answered Feb 14 '19 at 06:44

Use parameter engine='python' for change default engine='numexpr':

print(mpg.query('cylinders.isna()', engine='python'))

Sample:

mpg = pd.DataFrame({'cylinders':['a', np.nan]})
print(mpg)
  cylinders
0         a
1       NaN

print(mpg.query('cylinders.isna()', engine='python'))
  cylinders
1       NaN

More information about query is in Dynamic Expression Evaluation in pandas using pd.eval().

How to insert an isna() clause in a Pandas query

1 Answers1