Fuses a learner with preprocessing methods provided by caret::preProcess. Before training the preprocessing will be performed and the preprocessing model will be stored. Before prediction the preprocessing model will transform the test data according to the trained model.
After being wrapped the learner will support missing values although this will only be the case if ppc.knnImpute, ppc.bagImpute or ppc.medianImpute is set to TRUE.
makePreprocWrapperCaret(learner, ...)(Learner | character(1))
The learner.
If you pass a string the learner will be created via makeLearner.
(any)
See caret::preProcess for parameters not listed above.
If you use them you might want to define them in the add.par.set so that they can be tuned.
Other wrapper:
makeBaggingWrapper(),
makeClassificationViaRegressionWrapper(),
makeConstantClassWrapper(),
makeCostSensClassifWrapper(),
makeCostSensRegrWrapper(),
makeDownsampleWrapper(),
makeDummyFeaturesWrapper(),
makeExtractFDAFeatsWrapper(),
makeFeatSelWrapper(),
makeFilterWrapper(),
makeImputeWrapper(),
makeMulticlassWrapper(),
makeMultilabelBinaryRelevanceWrapper(),
makeMultilabelClassifierChainsWrapper(),
makeMultilabelDBRWrapper(),
makeMultilabelNestedStackingWrapper(),
makeMultilabelStackingWrapper(),
makeOverBaggingWrapper(),
makePreprocWrapper(),
makeRemoveConstantFeaturesWrapper(),
makeSMOTEWrapper(),
makeTuneWrapper(),
makeUndersampleWrapper(),
makeWeightedClassesWrapper()