map.hdgsom: Map data to a supervised or unsupervised hdgsom object

In alecuba16/HDGSOM: High Dimensionality Growing Self-Organizing Maps

Description

Map a data matrix onto a trained hdgsom object.

Usage

  ## S3 method for class 'hdgsom'
  map.hdgsom(hdgsom_object, df, retaindata = FALSE, ...)

Arguments

hdgsom_object	a trained hdgsom map (either unsupervised, as obtained by train.hdgsom, or supervised, as returned by train_xy.hdgsom).
df	Data matrix, with rows corresponding to objects, and columns to the dimensions the hdgsom object was trained with
retaindata	if set to TRUE a copy of the mapped data (unscaled) will be added to the returned object.
...	not used.

Value

Returns a S3 object of type "hdgsom" containing:

nodes$position	the location of the nodes on the map.
nodes$codes	codes that were established during the training for each node and dimension of the data.
nodes$freq	how many times each node was the best matching node for the mapped matrix.
mapped$bmn	contains the best matching node for each of the data that was mapped.
mapped$dist	distance from best matching node for each row of the mapped data-matrix.
data	Unscaled copy of the data that was mapped.

Author(s)

Alex Hunziker Alejandro Blanco Martinez

See Also

train.hdgsom, predict.hdgsom

Examples

  #Get some data
  data(iris)
  s = sample(1:150, 100)
  train_set = iris[s,1:4]
  test_set = iris[-s,1:4]

  # Create a GSOM Model
  hdgsom_iris <- train.hdgsom(train_set, spreadfactor=0.75)

  # Mapping
  mapped_iris <- map.hdgsom(hdgsom_iris, test_set)

alecuba16/HDGSOM documentation built on April 11, 2020, 5:06 a.m.