Distance funtion of knn classifier

1 次查看(过去 30 天)
Dear,none
I used Bayesian Optimization to choose the best parameters for a knn classifier.
After optimizing I got a knn with JACCARD distance and INVERSE distance weights, and no standardization.
I suppose this means that the nearest points are those that are the most distant (when using the jaccard).
Is anyone able to tell me how may I reproduce the INVERSE jaccard (with no standardization) by using the KDTreeSearcher or the ExhaustiveSearcher??
In other words, I would like to find, for each point in a test set, all the points that are the nearest, according to the distance function used by the optimized knn classifier.
THANKS a lot for any help!

回答(0 个)

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by