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
@carlzimmerman Brought up the fact that the nearest neighbor search used for KNNClassifier and KNNRegression could be sped up. Currently it does a full scan of all points in the training set.
Some potential options for speeding this up is to store training data in a KD tree or a ball tree.
@carlzimmerman Brought up the fact that the nearest neighbor search used for
KNNClassifier
andKNNRegression
could be sped up. Currently it does a full scan of all points in the training set.Some potential options for speeding this up is to store training data in a KD tree or a ball tree.
Introduction to KD trees:
https://www.analyticsvidhya.com/blog/2017/11/information-retrieval-using-kdtree/
Introduction to kd trees and ball trees:
https://ashokharnal.wordpress.com/tag/ball-tree-explained-in-simple-manner/
The text was updated successfully, but these errors were encountered: