|Machine Learning Added To Azure HDInsight|
|Written by Kay Ewbank|
|Friday, 13 July 2018|
Microsoft has announced the general availability of Machine Learning (ML) Services 9.3 on Azure HDInsight. The new release has added support for Python, hence the name change from R Server to ML Services.
ML Services on HDInsight can be used with either Python or R-based analytics on datasets loaded on either Azure Blob or Data Lake storage. As ML Services cluster is built on open-source R, it can be used with thousands of open-source R packages, along with routines from ScaleR, Microsoft’s big data analytics package. Microsoft says that:
"Any R or Python open-source machine learning package can work side by side with any proprietary innovation from Microsoft."
ML Services in HDInsight includes a highly scalable, distributed set of algorithms such as RevoscaleR, revoscalepy, and microsoftML that can work on data sizes larger than the size of physical memory.
ML Services comes with a set of pre-trained models for visual analysis and text sentiment analysis, ready to score data you provide. Once a machine learning model has been created, it can be turned into a web services API that is hosted on a server grid in the cloud and can be integrated with local applications.
The updated version of ML Services was announced alongside updates to Apache Hadoop, Apache Spark 2.3, Apache Kafka 1.0 and other bug fixes across the more than twenty open source frameworks that are part of HDInsight. HDInsight provides pre-tuned clusters that include the necessary products from the Hadoop ecosystem. The announcement also included a 50 percent price cut for HDInsight, a preview of HDInsight with Azure Data Lake Storage Gen2, and enhancements to the Enterprise Security Package. Azure Data Lake Storage Gen2 is a globally available HDFS filesystem to store and analyze petabyte-size files containing trillions of objects.
or email your comment to: email@example.com