pandas fill N.A. for specific column

Question

I want to fill N.A. values in a specific column if a condition is met in another column to only replace this single class of N.A. values with an imputed / replacement value.

E.g. I want to perform: if column1 = 'value1' AND column2 = N.A fillna_in_column2 with value 'replacementvalue'

How do I achieve this in pandas?

Trying to do this via dataframe[dataframe['firstColumn'] == 'value1'].fillna({'column2':'replacementValue'} does not work as the length of the overall records is modified. So far I could not get an inplace modification to work.

score 4 · Accepted Answer · answered Aug 29 '16 at 11:37

4

df.loc[(df['col1']==val1) & (df['col2'].isnull()), 'col2'] = replacement_value

answered Aug 29 '16 at 11:37

ayhan

70,170
20
182
203

gabra · Answer 2 · 2016-08-29T11:37:34.410

1

You can try this:

cond1 = df['column1'] == value1
cond2 = np.isnan(df['column2'])

df['column2'][cond1 & cond2] = replacement_value

edited Aug 29 '16 at 11:37

answered Aug 29 '16 at 11:32

gabra

9,484
4
29
45

pandas fill N.A. for specific column

2 Answers2