constructMBFLearner: Construct MoBaFeaS Gaussian Process Learner

In compstat-lmu/mobafeas: Model Based Feature Selection

Description

The mobafeas Learner is specifically set up to handle feature configuration vectors with a separate kernel.

Usage

constructMBFLearner(par.set, kernelconstructor = NULL,
  default.kernel = "matern3_2")

Arguments

par.set	[ParamSet] The parameter set of the objective function being optimized (probably the return value of makeMobafeasObjective()).
kernelconstructor	[function | NULL] function d -> kernel, see the Kernel functions.
default.kernel	[character(1)] One of "matern3_2" (default), "matern5_2", "exp", "gauss".

Value

Learner.

See Also

Other Surrogate Models: constructRFSurrogate

compstat-lmu/mobafeas documentation built on June 3, 2019, 7:19 a.m.