What's new in RapidMiner Studio 9.6
Switched bundled Java JRE in RapidMiner Studio from Oracle to OpenJDK
Changes in Oracle's JRE licensing and to be compliant with enterprise requirements we continue to move towards a free and up to date version of Java. With this release we moved to OpenJDK 8.
Creating connections of supported extensions is only possible using the new repository-based mechanism.
As a next step towards a more flexible way of managing data connectivity, we start to gradually reduce functionality of legacy mechanism. Creating new connection is only possible with the repository-based mechanism (see documentation here). Existing connections can still be edited and used, but this functionality will be removed eventually as well. Third party connections are not affected.
Added a Timezone parameter to JDBC connections
Solving timezone issues between your local environment and a database can be painful when retrieving data. The latest version of Studio extends the configurable parameters of JDBC connections with a timezone parameter to make sure that data is correctly retrieved from the connected database. See documentation for further information.
Improved Peak Detection and Peak Feature Extraction for Time Series data
New operators added to improve Time Series specific data transformation and feature extraction: * Highest Peak Transformation * Z-Score Peak Transformation * Extract Peaks * Equalize Numerical Indices * Equalize Time Stamps
Improved Universal Model Ops performance and functionality
- All custom prediction models are now supported, including grouped models
- Model Simulator supports grouped models as well, and uses raw input data and performs the operation
- Option to turn on/off explanation for scoring (disabling the options significantly speeds up scoring)
- Show if scores should be explained in overview table
- Optimized object loading for faster remote deployment
Enhancements and bug fixes
The following pages describe the enhancements and bug fixes in RapidMiner Studio 9.6 releases: