umap (version 0.2.2.0)

smooth.knn.dist: compute a "smooth" distance to the kth neighbor and approximate first neighbor

Description

compute a "smooth" distance to the kth neighbor and approximate first neighbor

Usage

smooth.knn.dist(k.dist, neighbors, iterations = 64,
  local.connectivity = 1, bandwidth = 1, tolerance = 1e-05,
  min.dist.scale = 0.001)

Arguments

k.dist

matrix with distances to k neighbors

neighbors

numeric, number of neighbors to approximate for

iterations

integers, number of iterations

local.connectivity

iteger

bandwidth

numeric

tolerance

numeric, for numeric precision

min.dist.scale

numeric, minimum distance to nearest neighbor (for display)

Value

list with two vectors, distances to the kth neighbor, and distance to the first nearest neighbor