The Dev Team at Revolution Analytics recently released an update to the Revolution R 5 family. Version 5.0.1 adds compatibility with Red Hat Enterprise Linux 6 for all editions (Community, Academic and Enterprise). This expands the platform support to Red Hat 5, Red Hat 6 and Microsoft Windows.
For Revolution R Enterprise customers and users of the free Academic edition, this update also addresses some minor issues with big-data statistics and distributed computing:
- The Revolution R Enterprise shortcut can now be pinned to your Start menu.
- K-means clustering (rxKmeans) with missing values was sometimes incorrectly reporting the number of valid and missing observations
- rxMerge was erroneously returning FALSE if the outfile argument had no .xdf extension, even though the merge was successful. Also, the use of the newVarNames argument sometimes led to the renaming of additional variables.
- rxSetVarInfo now returns an RxXdfData object if the inputdata is an .xdf file. If the input data is a data frame, a data frame is returned.
- rxFactors now returns an RxXdfData object if an output file is specified. Otherwise, a data frame is returned.
- The doRSR package now provides correct results when registered with a distributed compute context with wait=FALSE.
- RxHpcServer now behaves correctly if the names of all the compute nodes in a cluster are alphabetically after “Local Client.”
- The RevoDeployR web services framework has been updated (Academic and Server editions)
If you've previously downloaded Revolution R Community or the free Academic edition, you can download the latest version via our download page. If you're a current Revolution R Enterprise subscriber, you should already have an email with update instructions (if not, please contact Revolution Analytics support).
New to Revolution R? You can find information about our commercial-grade distribution of R at the link below.
Revolution Analytics: Revolution R Enterprise