The data mining pipeline is a submodule of sgpp::datadriven intended for out-of-the-box application of SG++ machine learning methods. By using the pipeline many algorithms of the datadriven module can be applied with minimal setup in C++. All boilerplate code is hidden from the user and configuration is done via JSON config files (see below).

The datamining pipeline is configured via JSON config files. Please refer to the configuration page for documentation of the configuration format, attributes and default values.