- The options
min_samples
andmax_samples
have been added to the values of the command line arguments--feature-sampling
and--instance-sampling
. - The indices of nominal and ordinal features must now be provided to a learner's
fit
-method via the keyword argumentsnominal_feature_indices
andordinal_feature_indices
. - The Python API does now allow to provide custom weights for training examples to a learner's
fit
-method via the keyword argumentsample_weights
. - Unnecessary conversions from integer weights to floating point values are now avoided.
- C++ 20 is now required for compiling the project.