BiocNeighborParam: The BiocNeighborParam class

In LTLA/kmknn: Nearest Neighbor Detection for Bioconductor Packages

The BiocNeighborParam class

Description

A virtual class for specifying the type of nearest-neighbor search algorithm and associated parameters.

Details

The BiocNeighborParam class is a virtual base class on which other parameter objects are built. There are currently 5 concrete subclasses in BiocNeighbors:

KmknnParam:

Exact nearest-neighbor search with the KMKNN algorithm.

VptreeParam:

Exact nearest-neighbor search with the tree algorithm.

ExhaustiveParam:

Exact nearest-neighbor search via brute-force.

AnnoyParam:

Approximate nearest-neighbor search with the Annoy algorithm.

HnswParam:

Approximate nearest-neighbor search with the HNSW algorithm.

These objects hold parameters specifying how each algorithm should be run on an arbitrary data set. See the associated documentation pages for more details.

Methods

In the following code snippets, x and object are BiocNeighborParam objects.

show(object):

Display the class and arguments of object.

bndistance(object):

Return a string specifying the distance metric to be used for searching. This should be one of "Euclidean", "Manhattan" or "Cosine".

x[[i]]:

Return the value of slot i, as used in the constructor for x.

x[[i]] <- value:

Set slot i to the specified value.

Author(s)

Aaron Lun

See Also

KmknnParam, VptreeParam, AnnoyParam, and HnswParam for constructors.

buildIndex, findKNN and queryKNN for dispatch.

LTLA/kmknn documentation built on Oct. 18, 2024, 9:01 p.m.