How to do log with 0 values in pandas dataframe

Question

How can I apply numpy log function to a panda dataframe that has 0 values in it?

This is an example of the series inside the dataframe I want to transform

df = pd.DataFrame({'CoapplicantIncome': [0, 1508.0, 0, 2358.0, 0], 'ApplicantIncome': [5849, 4583, 3000, 2583, 6000]}) ApplicantIncome CoapplicantIncome 0 5849 0.0 1 4583 1508.0 2 3000 0.0 3 2583 2358.0 4 6000 0.0

I tried to use

df['CoapplicantIncome'] = np.log(df['CoapplicantIncome'])

but it says 'cannot convert float NaN to integer'

After I do some searching, I've come to this line of code that says it will handle the 0 values.

df['CoapplicantIncome'] = np.log(df['CoapplicantIncome'].replace(0, np.nan))

But it still says 'cannot convert float NaN to integer'. So how can I solve this?

Interestingly, if I build this dataframe from scratch using pd.DataFramethe problem does not occur, but if I read the dataframe from a csv file, then the problem occur.

Same pandas version as me. Your issue isn't reproducible from your example. Presumably you have a NaN in there — roganjosh
– roganjosh, Commented Jul 21, 2018 at 11:47
@RafaelC ah right, cheers, I updated my pandas and it works now. Thanks — Iqbal Pratama
– Iqbal Pratama, Commented Jul 21, 2018 at 11:56

Iqbal Pratama · Accepted Answer · 2018-07-21 11:57:07Z

2

Fixed this problem by updating my pandas version.

answered Jul 21, 2018 at 11:57

Iqbal Pratama

1391 gold badge5 silver badges15 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

How to do log with 0 values in pandas dataframe

1 Answer 1

Comments

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Related