I was just wondering if anyone has been able to successfully use rapids with knn imputation. I know cuml.impute was avaliable perviously but is seems like it has not been removed. If anyone has a suggestion that would be great.
I tried using cuml.impute but it didn't work.
Just an update for anyone else, I have received a response from the cuml people and they basically said it is not a priority right now but any user made solutions are welcome.