cuda_ml_dbscan: Run the DBSCAN clustering algorithm.

In cuda.ml: R Interface for the RAPIDS cuML Suite of Libraries

Description

Run the DBSCAN (Density-based spatial clustering of applications with noise) clustering algorithm.

Usage

cuda_ml_dbscan(
  x,
  min_pts,
  eps,
  cuML_log_level = c("off", "critical", "error", "warn", "info", "debug", "trace")
)

Arguments

x	The input matrix or dataframe. Each data point should be a row and should consist of numeric values only.
min_pts, eps	A point 'p' is a core point if at least 'min_pts' are within distance 'eps' from it.
cuML_log_level	Log level within cuML library functions. Must be one of "off", "critical", "error", "warn", "info", "debug", "trace". Default: off.

Value

A list containing the cluster assignments of all data points. A data point not belonging to any cluster (i.e., "noise") will have NA its cluster assignment.

Examples

library(cuda.ml)
library(magrittr)

gen_pts <- function() {
  centroids <- list(c(1000, 1000), c(-1000, -1000), c(-1000, 1000))

  pts <- centroids %>%
    purrr::map(~ MASS::mvrnorm(10, mu = .x, Sigma = diag(2)))

  rlang::exec(rbind, !!!pts)
}

m <- gen_pts()
clusters <- cuda_ml_dbscan(m, min_pts = 5, eps = 3)

print(clusters)

cuda.ml documentation built on Jan. 8, 2022, 9:06 a.m.