Note
Click here to download the full example code
Mixed Nulls#
This notebook provides an overview for using and understanding the Mixed Nulls check.
Structure:
What are Mixed Nulls#
The MixedNulls
check search for various types of null values, including string representations of null.
Imports#
import pandas as pd
from deepchecks.tabular.checks import MixedNulls
Generate data#
data = {'col1': ['sog', '1', 'cat', None, None]}
dataframe = pd.DataFrame(data=data)
Run the check#
MixedNulls().run(dataframe)
We can also check for string representations of null:
data = {'col1': ['nan', None, 'null', 'Nan', '1', 'cat'], 'col2':['', '', 'None', 'a', 'b', 'c'], 'col3': [1,2,3,4,5,6]}
dataframe = pd.DataFrame(data=data)
MixedNulls().run(dataframe)
Total running time of the script: ( 0 minutes 0.170 seconds)