This post was authored by Nagesh Pabbisetty, Partner Director of Program Management, Microsoft Machine Learning Services.
Recently, at Strata Data Conference in New York City, Microsoft and Anaconda announced an exciting partnership to make Anaconda Python distribution into SQL Server, Machine Learning Server, Azure Machine Learning, and Visual Studio to deliver real-time insights. In addition, Anaconda will be distributing Microsoft R. Let’s take a deeper look at this exciting new partnership.
Microsoft is committed to helping developers build AI powered applications by enabling them to do machine learning and AI wherever their data is. SQL Server 2017 includes Machine Learning Services — enterprise grade in-database machine learning capabilities with R and Python languages. Machine Learning Server enables customers to do scalable machine learning using R or Python on standalone Windows and Linux servers, Hadoop clusters and Azure data platforms.
Anaconda is the leading distribution of Python leveraged by millions of users today. A strong partnership with this popular Python distribution for data science further strengthens Microsoft’s goal of building tools to empower every organization to build their own AI capabilities.
Microsoft and Anaconda built a customized Anaconda distribution – Anaconda for Microsoft for doing machine learning with Microsoft products and services. Packages from this distribution will initially be included in SQL Server 2017, Machine Learning Server and Azure Machine Learning.
As part of in-database Python support in SQL Server 2017 and Machine Learning Server, Microsoft makes packages from this distribution available as part of the setup to support data science and related functions. This helps customers get started with Python right out of the box without requiring them to install commonly needed packages separately. Data scientists have the continuity of using the familiar Anaconda distribution for doing in-database Python based machine learning in SQL Server without having to move data out of the secure and compliant SQL Server machines.
This partnership between Anaconda and Microsoft extends beyond Python – Anaconda will distribute Microsoft R as a part of its distribution. Anaconda users who use R language will get Microsoft R packages by default as part of Anaconda’s R Essentials package. This includes both Microsoft R Open and the proprietary R packages included in Microsoft R Client. Anaconda will use CRAN snapshots at Microsoft R Archived Network (MRAN) for building all the R packages available on Anaconda repositories.
Additionally, the Anaconda Navigator UI tool that helps orient the data scientists to the Anaconda ecosystem will include links to R Tools for Visual Studio for users to get started with R development using Visual Studio. Anaconda distribution will also include Visual Studio Code.
We are looking forward to growing this partnership that enables data scientists in enterprises to be highly productive with R and Python and build powerful enterprise AI applications. We encourage the readers to try out SQL Server 2017 and Machine Learning Server and explore the latest machine learning innovations.