Finds the k nearest neighbours for every point in a given dataset in O(N log N) time using Arya and Mount's ANN library (v1.1.3). There is support for approximate as well as exact searches, fixed radius searches and bd as well as kd trees.
This package implements nearest neighbors for the Euclidean (L2) metric. For the Manhattan (L1) metric, install the RANN1 package.
For further details on the underlying ANN library, see https://www.cs.umd.edu/~mount/ANN.
The recommendation is to install the released version from CRAN by doing:
install.packages("RANN")You can, however, use the remotes package to install the development version:
# install.packages("remotes")
remotes::install_github("jefferis/RANN")Please feel free to:
- submit suggestions and bug-reports at: https://github.com/jefferislab/RANN/issues
- send pull requests after forking: https://github.com/jefferislab/RANN/
- e-mail the maintainer: [email protected]
see inst/COPYRIGHT and DESCRIPTION files for copyright and license information.