error.quantisation: Quantisation error for a Self-Organising Map

In yasomi: Yet Another Self Organising Map Implementation

Description

error.quantisation is a generic function for computing the quantisation error made by a fitted Self-Organising Map on a dataset. The exact definition depends on the data type, but should be compatible with the standard definition for vector data; in this case the quantisation error is the mean of the Euclidean distance between each observation and the prototype of its best matching unit.

Usage

error.quantisation(som, newdata, ...)

Arguments

som	an object of class derived from "som"
newdata	optional observations (see details)
...	additional arguments to be passed to methods

Value

If newdata is not given, the function returns the quantisation error made by the fitted som on the data used to fit it.

When newdata is specified, the function returns the quantisation error of the fitted som on the corresponding data. The representation of the new data depends on the actual SOM.

Author(s)

Fabrice Rossi

See Also

error.kaskilagus, som.tunecontrol, error.quantisation.somnum for numerical SOM and error.quantisation.relationalsom for relational SOM

yasomi documentation built on May 2, 2019, 5:59 p.m.