What's new in RapidMiner Studio 9.10
Fit data with custom functions
Fit your data using any custom parameterized function using least squares. This is especially useful in industrial cases where some information is already known expected relations between attributes.
Some possible use cases:
- Model physical behaviour based on data
- Confirm physical hypothesis on measured data
- Anomaly detection on device behavior
- ...and many others
If the use of a specific column is likely to add unwanted bias to your models, it is highlighted as such. This happens in various places such as in the Statistics view of data, the model simulator, in Turbo Prep, in Auto Model, during model training, in model annotations among others.
Improvements in in-database extension
Breakpoints inside the nest allow for better and easier development. "Replace Missing Values" and "Declare Missing Value" operators have been added.
Time Series: providing better performance and reliability.
Code base in time series is now using Belt.
New feature for Equalize Indices Operators (Numerical Indices, Time Stamps) to select a constant replacement value. This allows to fill gaps in non-equalized data sets with constant values.
Direct access to the common industry automation platform OPC-UA with RapidMiner.
New operators to read and monitor industry asset data and browse to discover assets in a system.
Enhancements and bug fixes
The following pages describe the enhancements and bug fixes in RapidMiner Studio 9.10 releases. Note that 9.10.2 - 9.10.5 were not released for Studio: