Return to page


New versions of H2O-3 and Sparkling Water available


By Team | minute read | December 02, 2017

Blog decorative banner image

Dear H2O Community,
#H2OWorld  is on Monday and we can’t wait to see you there! We’ll also be live streaming the event  starting at 9:25am PST. Explore the agenda here .
Today we’re excited to share that new versions of H2O-3 and Sparkling Water are available.
We invite you to download them here: 

– MOJOs are now supported for Stacked Ensembles.
– Easily specify the meta-learner algorithm type that Stacked Ensemble should use. This can be AUTO, GLM, GBM, DRF or Deep Learning .
– GBM, DRF now support custom evaluation metrics.
– The AutoML leaderboard now uses cross-validation  metrics (new default).
– Multiclass stacking is now supported in AutoML . Removed the check that caused AutoML to skip stacking for multiclass.
– The Aggregator Function is now exposed in the Python/R client.
– Support for Python 3.6.
Detailed changes and bug fixes can be found here: 
Sparkling Water 2.0, 2.1, 2.2 
– Support for H2O Models into Spark python pipelines.
– Improved handling of sparse vectors  in internal cluster.
– Improved stability of external cluster deployment mode.
– Includes latest H2O-
Detailed changes and bug fixes can be explored here:
2.2 – 
2.1 – 
2.0 – 
Hope to see you on Monday!
The Team

 headshot Team

At, democratizing AI isn’t just an idea. It’s a movement. And that means that it requires action. We started out as a group of like minded individuals in the open source community, collectively driven by the idea that there should be freedom around the creation and use of AI.

Today we have evolved into a global company built by people from a variety of different backgrounds and skill sets, all driven to be part of something greater than ourselves. Our partnerships now extend beyond the open-source community to include business customers, academia, and non-profit organizations.