What is the new features of Spark 2.2? What all new API and updates has been updated to enhance machine learning in Spark 2.2?
Apache Spark 2.2 comes with big boost to machine learning with many new API and updated features.
Here are updates for MLlib, SparkR, and Python:
The latest version of Spark is here to support machine learning in a big way. It offers advanced analytics and python support. New things are:
Support for Locality Sensitive Hashing
Now it support Multiclass Logistic Regression
Also there is support for Personalized PageRank
Apache Spark 2.2 can be used with R to perform machine learning over distributed Spark cluster. In SparkR following features are added:
Now Spark 2.2 comes with the support of distribute machine learning over spark cluster. R programming code can be used to run over spark cluster.
Check Spark 2.2 Features.