RapidMiner 7.5 performance is fully upgraded !

 

What's new in the new version: increased performance for both in-memory and Hadoop workloads, new and improved cloud operators, prettier graphs, and much more.

Here’s the full list of what’s new:


 1. Refer a friend, get more data

  • Know someone who should be using RapidMiner Studio? RapidMiner Studio Free users can now refer a friend and get up to 10,000 additional rows of data for each referral.

  • To receive the additional rows of data, your friend must download RapidMiner Studio and create an account. RapidMiner Studio Free users can refer up to 4 friends,
    for a total of 50,000 rows of data.


 2. In-memory performance improvements                 

  • Lower memory footprint through more compact data representations.

  • Sparse datasets are detected and dealt with for an increase in efficiency.

  • The data core now keeps data in a columnar format，which significantly speeds up many data preparation tasks.


3. Hadoop performance improvements                 

  • A new mechanism has been implemented to manage Spark containers for in-Hadoop data preparation (using Hive-on-Spark) with greatly reduced latency.

  • You can now run multiple parallel jobs in a Hadoop cluster using the new parallel Loop and Loop Attributes operators available with RapidMiner Radoop.


 4. Improved integration with popular cloud platforms

  • New operators have been added for reading and writing data to/from Microsoft’s Azure Blob Storage.

  • • The Read Amazon S3 and Write Amazon S3 operators now support KMS encryption.


 5. New product tutorials

  • There are two new tutorials for RapidMiner Server and RapidMiner Radoop.

  • We’ve added a new mechanism to provide help and important announcements to the user directly inside RapidMiner Studio.


6. Improved graphs

  • We’ve revised all graph visualizations in the results view, including Decision Tree graphs, clustering, association rules, and a few more.


7. Better machine learning in Hadoop                 

  • The SparkRM operator now includes the Bootstrapping algorithm to easily create ensemble models including Random Forest.


8. Increased security with RapidMiner Radoop

  • RapidMiner Radoop now supports the option to use different databases for UDFs and for temporary tables.


Go and download the basic version of RapidMiner for a trial~