Properties can be accessed with getLearnerProperties(learner), which returns a
character vector.
The learner properties are defined as follows:
- numerics, factors, ordered
Can numeric, factor or ordered factor features be handled?
- functionals
Can an arbitrary number of functional features be handled?
- single.functional
Can exactly one functional feature be handled?
- missings
Can missing values in features be handled?
- weights
Can observations be weighted during fitting?
- oneclas, twoclass, multiclass
Only for classif: Can one-class, two-class or multi-class classification problems be handled?
- class.weights
Only for classif: Can class weights be handled?
- rcens, lcens, icens
Only for surv: Can right, left, or interval censored data be handled?
- prob
For classif, cluster, multilabel, surv: Can probabilites be predicted?
- se
Only for regr: Can standard errors be predicted?
- oobpreds
Only for classif, regr and surv: Can out of bag predictions be extracted from the trained model?
- featimp
For classif, regr, surv: Does the model support extracting information on feature importance?
getLearnerProperties(learner) hasLearnerProperties(learner, props)
Arguments
| learner | (Learner | |
|---|---|
| props | (character) |
Value
getLearnerProperties returns a character vector with learner properties.
hasLearnerProperties returns a logical vector of the same length as props.
See also
Other learner:
getClassWeightParam(),
getHyperPars(),
getLearnerId(),
getLearnerNote(),
getLearnerPackages(),
getLearnerParVals(),
getLearnerParamSet(),
getLearnerPredictType(),
getLearnerShortName(),
getLearnerType(),
getParamSet(),
helpLearnerParam(),
helpLearner(),
makeLearners(),
makeLearner(),
removeHyperPars(),
setHyperPars(),
setId(),
setLearnerId(),
setPredictThreshold(),
setPredictType()