nnetlearners to mlr3learners
classif.lightgbmrespectively. Copied from mlr3learners.lightgbm
LearnerClassifLiblineaRXdeprecated in favour of only two learners (
LearnerClassLiblineaR) with added hyper-parameters. Deprecated learners will be removed in v0.3.0.
classif.nnetwill be removed in v0.4.0.
liblinearXwill be removed in v0.4.0.
dist = "logistic"has been removed from
surv.parametricas it is unclear what this was previously predicting.
type = "tobit"for
dist = "gaussian"so predictions can correspond with
LearnerRegrGlmwith the unique key
stats, which allows users to change the
familyhyperparameter when fitting generalized linear regression models.
classif.bartas this is forced internally
sepredict type to
install_learnersnow additionally installs required mlr3 packages
surv.parametricoccurring if feature names are switched between training and predicting
classif.nnet, in the future please load from mlr3learners
survlearners that were reversing the order of
crank, see this issue for full details: https://github.com/mlr-org/mlr3proba/issues/165
responseis no longer returned by
survivalmodelslearners which weren’t returning
surv.coxboost_cvcan now only handle
numericfeature types, previous automated internal coercions were inconsistent with mlr3 design.