Why is SimpleImputer's fit_transform not working for dataframe in google colab?

Question

imp = SimpleImputer(missing_values=np.nan, strategy='most_frequent')
weather_test = imp.fit_transform(weather_test)

The above code is throwing error in google colab when weather_test is a pandas dataframe. But when I change weather_test to numpy array, it works.

imp = SimpleImputer(missing_values=np.nan, strategy='most_frequent', verbose=0)
weather_test = imp.fit_transform(np.array(weather_test))

Do you mind to add the full error? – rpanai Jan 09 '20 at 14:02 — rpanai, Jan 09 '20 at 14:02

score 0 · Answer 1 · answered Jan 09 '20 at 14:02

Scikit-learn's API methods generally assume the input will be a numpy array rather than a pandas dataframe. For an example of how to use this functionality with a dataframe, see How to use sklearn fit_transform with pandas and return dataframe instead of numpy array?

1 Answers1