Major new features in this release includes:
- xgboost (v0.90) support (find usage)
- Improvements on tokenize_ja
- libsvm format converter is supported (find usage)
- Improve support for Json conversion (
to_json/from_json
) (find usage) - Introduced tracing functionality for DecisionTree predictions (find usage)
- Support for
max_by
,min_by
,majority_vote
UDAFs (find usage) - Supoprt
argmax/argmin
andargsort/argrank
UDFs (find usage) - Refined RandomForest implementation for Sparse feature support and training speed. (find usage)
Please refer ChangeLog for the detail of changes.
Major updated in this release includes:
- Supported Apache Spark v2.3.
- Brickhouse UDFs
- Field-aware Factorization Machines
Please refer ChangeLog for the detail of changes.
The first Apache release.
Please Refer ChangeLog for the detail of changes.