som.nn (version 1.4.4)

som.nn.som.experimental: Work hourse for som training.

Description

Function is the kernel internal for som training, implemented in pure R.

Usage

som.nn.som.experimental(
  data,
  grid,
  len = 100,
  alpha = 0.05,
  radius,
  init,
  toroidal = FALSE
)

Value

    S3 object of type \code{kohonen} with the trained som.

Arguments

data

matrix with training data.

grid

somgrid object

len

number of steps to be trained (steps - not epochs!).

alpha

learning rate c(first, last).

radius

radius c(first, last).

init

codes for initialisation.

toroidal

true if doughnut-shaped som.