« Changelog History


H2O v3.26.0.2 Release Notes

  • 🚀 Download at: http://h2o-release.s3.amazonaws.com/h2o/rel-yau/2/index.html

    Bug

    🛠 [PUBDEV-6112] - Fixed an NPE error that occurred on models StackedEnsemble in AutoML. [PUBDEV-6587] - Improve the error message for rbind failures that resulted when rbinding datasets with long categorical levels. [PUBDEV-6693] - In Flow, the scoring history deviance graph no longer displays if a custom distribution is not set. [PUBDEV-6709] - pr_auc() now works correctly in the Python client.

    New Feature

    👍 [PUBDEV-6255] - Added support for Target Encoding MOJOs. 👍 [PUBDEV-6593] - Added support for Target Encoding transformation of data without a response column. [PUBDEV-6640] - Added TargetEncoderBuilder (estimator) and TargetEncoderModel (transformer). [PUBDEV-6682] - Added detailed MOJO metrics for DRF, Isolation Forest, and GLM MOJO models. [PUBDEV-6684] - Added AUCPR to the list of available stopping_metric options.

    Improvement

    [PUBDEV-6436] - In Flow, users can now upload a MOJO, and a generic model will automatically be created from it. 🚚 [PUBDEV-6681] - Removed duplicated code for obtaining logs in Java. 🔊 [PUBDEV-6685] - Improved error handling in the downloadLogs method. [PUBDEV-6690] - Disabled autocomplete on the Flow login form.

    📄 Docs

    [PUBDEV-6674] - Added an entry for upload_custom_metric in the Parameters Appendix. 🏗 [PUBDEV-6678] - Added list of parameters that can be specified when building a Generic Model (MOJO import). 📚 [PUBDEV-6687] - Updated documentation for MOJO Import. [PUBDEV-6698] - Added "aucpr" to the list of available stopping_metric options. [PUBDEV-6706] - Added an entry for export_checkpoints_dir in the Parameters Appendix.