CODEWITHSUNDEEP
pythonpandasdataframe
David H
David H

Reputation: 5

How to replace Nan and negative number with zero

data frame

In the above dataframe I am trying to replace NaN and negative values in the last column with 0 using pandas, I have used some of the suggestions here, but it also seems to affect other columns and replace values in the other columns with zeroes.

How do I restrict it to just the selected column

Upvotes: 0

Views: 3047

Answers (2)

SeaBean
SeaBean

Reputation: 23217

You can use .mask() to change negative numbers to NaN and then fillna() to 0 together with other NaN values in one go, as follows:

df['New_expenses'] = df['New_expenses'].mask(df['New_expenses'] < 0).fillna(0)

Or, even more simpler, credit to @tdy, using .where() instead:

df['New_expenses'] = df['New_expenses'].where(df['New_expenses'] >= 0, 0)

.where() keeps the values when the condition is true and replaces values when the condition is false.

In this case, we keeps the values when the values are positive and replaces the values to 0 when the condition is false (including negative numbers and NaN values).

Upvotes: 3

mozway
mozway

Reputation: 260600

Use clip to set the negative values to a minimum, and fillna to replace the NAs:

df['New_expenses'] = df['New_expenses'].clip(lower=0).fillna(0)

Upvotes: 2

Related Questions