CODEWITHSUNDEEP
pythonpandasnan
zavy mola
zavy mola

Reputation: 143

df.fillna : Gives Error because of replace value for NaN not present in categorical attributes

df.fillna('unknown',inplace=True)

I have a dataframe df wherein some attributes are categorical and some are objects. Using the above expression I am trying to fill missing values in any of these columns with 'unknown'. But I get an erro while doing so saying 'unknown' is not present in categories of the categorical data columns.

How can I do it?

Upvotes: 1

Views: 260

Answers (1)

Chris Adams
Chris Adams

Reputation: 18647

Use DataFrame.select_dtypes and loop through each categorical column and add 'unknown' as a category using add_categories method on the .cat accessor.

for col in df.select_dtypes('category'):
    df[col].cat.add_categories('unknown', inplace=True)

df.fillna('unknown',inplace=True)

Upvotes: 1

Related Questions