Package elki.index.tree.spatial.kd.split
Class MeanVarianceSplit
- java.lang.Object
-
- elki.index.tree.spatial.kd.split.MeanVarianceSplit
-
- All Implemented Interfaces:
SplitStrategy
@Reference(authors="S. M. Omohundro", title="Efficient Algorithms with Neural Network Behaviour", booktitle="Journal of Complex Systems 1(2)", url="https://www.complex-systems.com/abstracts/v01_i02_a04/", bibkey="journals/jcs/Omohundro87") public class MeanVarianceSplit extends java.lang.Object implements SplitStrategy
Split on the median of the axis with the largest variance.Reference:
S. M. Omohundro
Efficient Algorithms with Neural Network Behaviour
Journal of Complex Systems 1(2)- Since:
- 0.8.0
- Author:
- Erich Schubert
-
-
Nested Class Summary
Nested Classes Modifier and Type Class Description static class
MeanVarianceSplit.Par
Parameterizer-
Nested classes/interfaces inherited from interface elki.index.tree.spatial.kd.split.SplitStrategy
SplitStrategy.Info, SplitStrategy.Util
-
-
Field Summary
Fields Modifier and Type Field Description static MeanVarianceSplit
STATIC
Static instance.
-
Constructor Summary
Constructors Constructor Description MeanVarianceSplit()
-
Method Summary
All Methods Instance Methods Concrete Methods Modifier and Type Method Description SplitStrategy.Info
findSplit(Relation<? extends NumberVector> relation, int dims, ArrayModifiableDBIDs sorted, DBIDArrayMIter iter, int left, int right, VectorUtil.SortDBIDsBySingleDimension comp)
Build the k-d-tree using midpoint splitting.
-
-
-
Field Detail
-
STATIC
public static final MeanVarianceSplit STATIC
Static instance.
-
-
Method Detail
-
findSplit
public SplitStrategy.Info findSplit(Relation<? extends NumberVector> relation, int dims, ArrayModifiableDBIDs sorted, DBIDArrayMIter iter, int left, int right, VectorUtil.SortDBIDsBySingleDimension comp)
Description copied from interface:SplitStrategy
Build the k-d-tree using midpoint splitting.- Specified by:
findSplit
in interfaceSplitStrategy
- Parameters:
relation
- data relationdims
- data dimensionalitysorted
- modifiable array to encode the partitionsiter
- reusable iterator to the arrayleft
- left subintervalright
- right subintervalcomp
- reusable comparator (bound to the relation above)- Returns:
- split info, null if no good split
-
-