mlr3learners 0.8.0

• fix: Hyperparameter set of lrn("classif.ranger") and lrn("regr.ranger").
  Remove alpha and minprop hyperparameter.
  Remove default of respect.unordered.factors.
  Change lower bound of max_depth from 0 to 1.
  Remove se.method from lrn("classif.ranger").
• feat: use base_margin in xgboost learners (#205).
• fix: validation for learner lrn("regr.xgboost") now works properly. Previously the training data was used.
• feat: add weights for logistic regression again, which were incorrectly removed in a previous release (#265).
• BREAKING CHANGE: When using internal tuning for xgboost learners, the eval_metric must now be set.
  This achieves that one needs to make the conscious decision which performance metric to use for early stopping.
• BREAKING CHANGE: Change xgboost default nrounds from 1 to 1000.

mlr3learners 0.7.0

• feat: LearnerClassifXgboost and LearnerRegrXgboost now support internal tuning and validation.
  This now also works in conjunction with mlr3pipelines.

mlr3learners 0.6.0

• Adaption to new paradox version 1.0.0.

mlr3learners 0.5.8

• Adaption to memory optimization in mlr3 0.17.1.

mlr3learners 0.5.7

• Added labels to learners.
• Added formula argument to nnet learner and support feature type "integer"
• Added min.bucket parameter to classif.ranger and regr.ranger.

mlr3learners 0.5.6

• Enable new early stopping mechanism for xgboost.
• Improved documentation.
• fix: unloading mlr3learners removes learners from dictionary.

mlr3learners 0.5.4

• Added regr.nnet learner.
• Removed the option to use weights in classif.log_reg.
• Added default_values() function for ranger and svm learners.
• Improved documentation.

mlr3learners 0.5.2

• Most learners now reorder the columns in the predict task according to the
  order of columns in the training task.
• Removed workaround for old mlr3 versions.

mlr3learners 0.5.1

• eval_metric() is now explicitly set for xgboost learners to silence a
  deprecation warning.
• Improved how the added hyperparameter mtry.ratio is converted to mtry to
  simplify tuning.
• Multiple updates to hyperparameter sets.

mlr3learners 0.5.0

• Fixed the internal encoding of the positive class for classification learners
  based on glm and glmnet (#199). While predictions in previous versions
  were correct, the estimated coefficients had the wrong sign.
• Reworked handling of lambda and s for glmnet learners (#197).
• Learners based on glmnet now support to extract selected features (#200).
• Learners based on kknn now raise an exception if k >= n (#191).
• Learners based on ranger now come with a virtual hyperparameter mtry.ratio
  to set the hyperparameter mtry based on the proportion of features to use.
• Multiple learners now support the extraction of the log-likelihood (via method
  $loglik(), allowing to calculate measures like AIC or BIC in mlr3 (#182).