Using kNN to impute missing data with sklearn

Question

I took a look at this question here: Missing value imputation in python using KNN

I don't believe I can use fancyimpute or download the sklearn knn impute code from github (I'm doing this on a python platform where I can't just download additional code).

Would there be a way to use sklearn's kNN to do this manually? If I have data missing for some columns I don't think sklearn's kNN will run right?

score -2 · Answer 1 · answered Apr 02 '18 at 08:38

-2

Answering your last question:

Missing data will mean that the distance between samples can not be calculated. These samples could be excluded. So if you have missing data, sklearn's knn will not run right.

answered Apr 02 '18 at 08:38

Ashutosh Chapagain

930
9
15

this should have been a comment on the original post but not enough reputation. I dont think there is a method within knn without additional downloads or fancyimpute to fill up missing values. – Ashutosh Chapagain Apr 02 '18 at 08:40

Using kNN to impute missing data with sklearn

1 Answers1