September 25th, 2019

H2O Driverless AI Acceleration with Intel DAAL

RSS icon RSS Category: Data Science, Driverless AI, Machine Learning
Fallback Featured Image

This week at Strata NY 2019 we will be demoing a custom recipe that incorporates the Intel Data Analytics Acceleration Libraray (DAAL) algorithm into Driverless AI. This blog will provide an introduction to Intel DAAL and how the Make-Your-Own-Recipe capability extends H2O Driverless AI.  If you are at Strata NY 2019, stop by the Intel booth for a demo.

H2O.ai and Intel: Democratizing AI for Everyone

H2O.ai and Intel created ProjectBlue Danube to collaborate on accelerating machine learning algorithms and libraries on Intel platforms including the new 2nd Gen Intel Xeon Scalable processors and Intel® Optane™ DC persistent memory, announced last April, which were developed to deliver agility, scale and security for AI workloads. The next phase of Project Blue Danube, enables the world’s leading enterprises to create highly scalable, high performance, more secure and accelerated data science workflows on the world’s most pervasive platform. The combination allows enterprise organizations to gain a competitive edge by becoming AI companies with a highly scalable, cost-effective and faster path to insights and results.  

Now with the new “Make Your Own AI” recipe capability available with H2O Driverless AI, Intel created and unveiled the DAAL recipe on the H2O.ai open source recipe repository to enable enterprises to achieve machine learning at speed and scale. The results that H2O.ai achieved on the latest Intel platform are impressive vs traditional systems.

AI to do AI

H2O Driverless AI empowers data scientists, data engineers, mathematicians, statisticians and domain scientists to work on projects faster and more efficiently by using automation to accomplish tasks that can take months and can now be reduced to hours or minutes by delivering automatic feature engineering, model validation, model tuning, model selection and deployment, machine learning interpretability, time-series, NLP and automatic pipeline generation for model scoring, and now bring your own recipes and model operations and administration.

As highlighted by Forrester Research, H2O.ai “Is best for companies that want to delight data science teams,” write Forrester Analysts Kjell Carlsson and Mike Gualtieri in The Forrester New Wave: Automation-Focused Machine Learning Solutions, Q2 2019.

Custom Recipe via Bring Your Own Recipe (BYOR)

H2O Driverless AI provides a new feature called “custom recipes”. These recipes are essentially custom snippets of code which can incorporate any machine learning algorithm, any scorer/metric and any feature transformer. A user can create custom recipes using python utilizing any external library or his/her own creations. This is an interesting feature because it allows Driverless AI’s automated approach to be enhanced by the data scientist’s expert domain knowledge on his/her field. If you’d like to understand more about DAI’s custom recipes and get hands on check out the following recipe tutorial.

As part of this, we have created a github repository where we share via open source some of our own custom recipes . There are different folders which cover different ML areas, for example NLP, Time Series, categorical features, numerical features, geospatial and so on.  

You can also find the Intel DAAL recipe github in repository.  Within a few clicks the new algorithm can be added to the Driverless AI env and ready to be considered in the model building optimization process. The following animated image walks you through the steps.

Learn more

To learn more and see a demo join our webinar with Intel on Oct 3rd.

About the Author

Rafael Coss

Rafael Coss is a Community and Partner Maker at H2O.ai. Prior to joining H2O.ai, he was technical marketing and community Director and a developer advocate at Hortonworks. He was also the DataWorks Summit Program Co-Chair for the past 3 years. Prior to Hortonworks he was a Senior Solution Architect and Manager of IBM’s WW Big Data Enablement team. At IBM he was responsible for the technical product enablement for BigInsights and Streams. Previously, he held several other positions in IBM, where he worked on tools, XML db, federated db and Object-Relational db.

Leave a Reply

Novel Ways To Use Driverless AI

I am biased when I write that Driverless AI is amazing, but what's more amazing

November 14, 2019 - by Thomas Ott
Useful Machine Learning Sessions from the H2O World New York

Conferences not only help us learn new skills but also enable us to build brand

November 13, 2019 - by Parul Pandey
Fallback Featured Image
Accelerate Machine Learning workflows with H2O.ai Driverless AI on Red Hat OpenShift, Enterprise Kubernetes Platform

Organizations globally are operationalizing containers and Kubernetes to accelerate Machine Learning lifecycles as these technologies

November 12, 2019 - by Nicholas Png
Image Tasks on H2O Driverless AI

I’d like to thank Grandmaster Yauhen Babakhin for reviewing the drafts and the very useful

November 12, 2019 - by Sanyam Bhutani
Importing, Inspecting, and Scoring With MOJO Models Inside H2O

Machine-learning models created with H2O may be exported in two basic ways: Binary format, Model Object, Optimized

November 8, 2019 - by Pavel Pscheidl
Natural Language Processing in H2O’s Driverless AI

Note: I’d like to thank Grandmaster SRK for a lot of suggestions and corrections with

November 6, 2019 - by Sanyam Bhutani

Join the AI Revolution

Subscribe, read the documentation, download or contact us.

Subscribe to the Newsletter

Start Your 21-Day Free Trial Today

Get It Now
Desktop img