RapidMiner 9.8 Beta is now available
Be one of the first to get your hands on the new features. More details and downloads here:
Accessing a process definition and metadata
I expected this to work with built-in RapidMiner operators, but neither Loop Repository nor Retrieve nor Open File will open a process from the repository.
However, with a simple Groovy script (in the built-in Execute Script operator) it's easy to get the process definition as XML, and then process it further with RapidMiner operators.
import com.rapidminer.repository.RepositoryLocation; import com.rapidminer.RepositoryProcessLocation; //Current repository path parentLoc = operator.getProcess().getRepositoryLocation().parent(); //Other repository entry, relative to the current path loc = new RepositoryLocation(parentLoc, "Example process"); //Process to read as XML processCode = new RepositoryProcessLocation(loc).getRawXML(); //Add a macro with the process XML operator.getProcess().getMacroHandler().addMacro("processXML", processCode);The result is a macro that contains the specified process formatted as XML. You can then use RapidMiner operators to process the data further and extract the things you want.
You can find more details and a downloadable example process in my blog on datascientist.at.