ModeSeekingVersion1: Mode Seeking Algorithm

In Epsilon127/ldder: Local Density Derivative Estimation

Description

Iterative mean shift algorithm detecting the mode of the underlaying pdf.

Usage

ModeSeekingVersion1(data, h, method = "LHSE", type = "Gaussian",
  startpoints = NULL, delta = 1e-10, iter_max = Inf)

Arguments

data	Data matrix [d, n].
h	Kernel bandwidth parameter
method	Used method for the log-density estimation. Either LHSE, LMME or KDE.
type	Used kernel. Either Gaussian or Epan-(echikov).
startpoints	Matrix [d, m] containing in each raw a starting point for the algorithm. If NULL the data points will be used.
delta	Stopping criteria.
iter_max	Maximal number of iterations before the procedure stops. Will print a warning if it did not converge.

Details

The algorithm uses LHSE, LMME or KDE for finding the modes of the underlaying pdf.

Value

Fixpoints of the iteration as matrix [d, m]

Epsilon127/ldder documentation built on March 7, 2020, 7:47 a.m.