buildIndex: Build a nearest-neighbor index

In LTLA/kmknn: Nearest Neighbor Detection for Bioconductor Packages

Build a nearest-neighbor index

Description

Build indices for nearest-neighbor searching with different algorithms.

Usage

buildIndex(X, ..., BNPARAM)

Arguments

X	A numeric matrix where rows correspond to data points and columns correspond to variables (i.e., dimensions).
...	Further arguments to be passed to individual methods. This is guaranteed to include transposed.
BNPARAM	A BiocNeighborParam object specifying the type of index to be constructed. This defaults to a KmknnParam object if no argument is supplied.

Details

Supplying a KmknnParam object as BNPARAM will dispatch to buildKmknn.

Supplying a VptreeParam object as BNPARAM will dispatch to buildVptree.

Supplying an AnnoyParam object as BNPARAM will dispatch to buildAnnoy.

Supplying an HnswParam object as BNPARAM will dispatch to buildHnsw.

Value

An instance of a BiocNeighborIndex subclass, containing indexing structures for the specified algorithm.

Author(s)

Aaron Lun

See Also

buildKmknn, buildVptree, buildAnnoy and buildHnsw for specific methods.

Examples

Y <- matrix(rnorm(100000), ncol=20)
(k.out <- buildIndex(Y))
(a.out <- buildIndex(Y, BNPARAM=AnnoyParam()))

LTLA/kmknn documentation built on Feb. 5, 2024, 6:03 p.m.