Meta Model Interface: Maximum Likelihood Estimation for Gaussian Processes, Kriging

Share:

Description

Kriging model based on mlegp package. This function uses two settings, which are stored in the spotConfig parameter:
spotConfig$seq.mlegp.constantMean Use constant mean (mu) in mlegp (=1) or linear model (=0); 1 by default
spotConfig$seq.mlegp.min.nugget minimum value of nugget term; 0 by default

If those settings are not in spotConfig their mentioned defaults will be used.

If the numeric value of spotConfig$mlegp.reduce is smaller than the observations in mergedB, spotConfig$mlegp.reduce will specify how many samples should be drawn without replacement from mergedB. This can prevent explosion of time consumption in this function. Mlegp can be used both for single and multi objective SPOT.

Usage

1
spotPredictMlegp(rawB, mergedB, design, spotConfig, fit = NULL)

Arguments

rawB

matrix of raw x and y values

mergedB

matrix of merged x and y values, does not have replicate entries

design

design points to be evaluated by the meta model

spotConfig

the list of all parameters is given.

fit

if an existing model fit is supplied, the model will not be build based on data, but only evaluated with the model fit (on the design data). To build the model, this parameter has to be NULL. If it is not NULL the parameters mergedB and rawB will not be used at all in the function.

Value

returns the list spotConfig with two new entries:
spotConfig$seq.modelFit fit of the model used with predict()
spotConfig$seq.largeDesignY the y values of the design, evaluated with the fit

See Also

SPOT

Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.