tunedagsvm_extractoptimisedparametersandmodels: Extracts the DAG-SVM of a set of tuned DAG-SVM with the Best...

In henningte/TraceIdentification: Identification of Fluorescent Traces in High-Resolution Aerial Images

Description

tunedagsvm_extractoptimisedparametersandmodels takes the output of tunedagsvm, i.e. a set of tuned directed acyclic graph-support vector machines (DAG-SVM) \insertCitePlatt.2000TraceIdentification for a specific classification task and returns for each node the SVM that had the maximum overall classification accuracy when running tunedagsvm.

Usage

tunedagsvm_extractoptimisedparametersandmodels(optimparam,
  kernel = "radial")

Arguments

optimparam	A list as returned by tunedagsvm containing performance parameters and tuned DAG-SVM.
kernel	A character value indicating a specific kernel for which to extract the tuned SVM for each node of the DAG-SVM. If kernel = NULL, tunedagsvm_extractoptimisedparametersandmodels select for each node the kernel with the maximum overall classification accuracy. kernel is intended to provide the possibility to select a specific desired kernel for a DAG-SVM.

Value

The function returns a list with two elements:

prunedoptimparam

A data.frame object containing for each node the respective selected row of the first element of optimparam.

prunedbestmodels

A list with an element for layer containing the SVM for all nodes of this layer. Each element of prunedbestmodels is a list of the respective SVM.

References

\insertAllCited

See Also

tunedagsvm.

Examples

#

henningte/TraceIdentification documentation built on May 26, 2019, 6:50 p.m.