You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Ensure that value K is accessible for numerical and ideally also categorical values
Try to find some strategy to impute categorical values with e.g. most frequent -> explore scanpy KNN (neighbors) for custom implementation if required. Likely will involve label encoding the values, imputing and then mapping back.
Google around for KNN imputation improvements
The text was updated successfully, but these errors were encountered:
- added n_neighbours parameter to knn impute
- encoding now only gets applied to non numerical columns
- added tests for knn impute with n_neighbours parameter
The text was updated successfully, but these errors were encountered: