The major improvement to version 4.2 is a totally new Data Matrix (DM) concept. This new DM format is now built into our tools and allows to handle big data. For example it is possible to make robustness evaluations for models trained in Brainer with huge data amounts directly in Stipulator.

Stipulator

• Extraction redesign: therefor a new synchronization function sssync() was born. This allows an easier model evaluation in Stipulator (because same functions in processing and extraction are used now).
• Several bugfixes and compatibility issues with new Matlab 2017b.

Brainer

• Brand new data import GUI.
• New model type Gaussian Process Regression for regression problems.

Mapper

• New model type selection and parametrisation user interfaces (using same dialogues and concept as in Brainer).
• Bugfix in calculation of best matching unit.

Expectator

• Expectations can be evaluated with only some available channels now (in former versions all defined channels must be available for evaluation).
• Extensions like in Stipulator are now available.
• Several bugfixes and compatibility issues with new Matlab 2017b.