Trees

mlpack includes a number of space partitioning trees and other trees for its geometric techniques. All of mlpack’s trees implement the same API, allowing easy plug-and-play usage of different trees. The following tree types are available in mlpack:

KDTree
MeanSplitKDTree
BallTree
MeanSplitBallTree
RPTree
MaxRPTree
BinarySpaceTree
UBTree

Note: this documentation is a work in progress. Not all trees are documented yet.

In general, it is not necessary to create an mlpack tree directly, but instead to simply specify the type of tree a particular algorithm should use via a template parameter. For instance, all of the algorithms below use mlpack trees and can have the type of tree specified via template parameters:

NeighborSearch (for k-nearest-neighbor and k-furthest-neighbor)
RangeSearch
KDE
FastMKS
DTB (for computing Euclidean minimum spanning trees)
KRANN

Note: if you are looking for documentation on decision trees, see the documentation for the DecisionTree class.