Kernel: Feature Selection Kernels

In compstat-lmu/mobafeas: Model Based Feature Selection

Description

Kernels defined on the space of possible feature selections. These should be run, possibly with arguments, and the result given to constructMBFLearner().

Usage

kernelMBFHamming()

kernelMBFGraph(allequal = TRUE)

kernelMBFAgreement(limit.par = FALSE)

kernelMBFAgreeCor(data, limit.par = FALSE)

Arguments

allequal	[logical(1)] Whether to generate one parameter per feature, or only one parameter overall.
limit.par	[logical(1)] Whether to limit the parameter value range
data	[matrix | data.frame | Task] the data to use.

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