We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
75ba481
Fixed max leaves bug, and make object pickleable.
60c3030
Adding more sklearn compatibility changes and methods to the python API.
91318f1
Default all metadata to use ObjectSerializer for better consistency.
c5010ae
Makes the python object cloneable using the scikit-learn API, and then refactors with a lot more tests and brings python test coverage to 100%.
aeeb3e2
Make it so that if there are no missing records in the missing node, it just get's the weight of the parent.
7cb2951
Added the feature_importances_ attribute to the python api, to use with sklearn feature selectors.
feature_importances_
b82bcb0
Fix bug with missing leaf not getting included in collected stats, impacting variable importance.
5c0d30e
Saved last tree when early stopping is hit, in addition raise a warning if multiple evaluation datasets are used.
ddb32f8
Fix issue where early stopping rounds wasn't actually stopping training.
145f705
Added the ability to force children weights to contain the parent weight, also fixed bug where learning rate wasn't applied to the root node.
