tdmDispatchTuner: Helper function for 'tdmBigLoop'.

In TDMR: Tuned Data Mining in R

Description

tdmDispatchTuner selects and starts the tuner specified by tuneMethod.
See the 'Details' section of tdmBigLoop for a list of available tuners.

Usage

tdmDispatchTuner(tuneMethod, confFile, spotStep, tdm, envT, dataObj)

Arguments

tuneMethod	the tuning algorithm given as a string. Possible values are { "spot" | "lhd" | "cmaes" | "cma_j" | "bfgs" }.
confFile	the configuration name.
spotStep	–DEPRECATED– SPOT 2.0 supports only spotStep="auto".
tdm	the TDMR object
envT	the environment variable
dataObj	the TDMdata object containing the data set (train/vali part and test part)

Value

The result tunerVal of the tuning algorithm, this is the list envT$spotConfig, extended by

alg.currentResult	the RES data frame
alg.currentBest	the BST data frame
...	[optional] further results from the tuning algorithm tuneMethod

Note

Side effects: Environment envT is extended by

• envT$tunerVal = tunerVal

• envT$res the RES data frame

• envT$bst the BST data frame

Author(s)

Wolfgang Konen, THK, 2018

See Also

tdmBigLoop

TDMR documentation built on March 3, 2020, 1:06 a.m.