RapidMiner 9.7 is Now Available

Lots of amazing new improvements including true version control! Learn more about what's new here.


Setting Macros

mdrmdr Member Posts: 7 Contributor II
edited November 2018 in Help

I am somewhat new to RM but not to ML. I have run some experiments so far, 10 fold et.al. with ID3, Naive Bayes, SVM, KNN.

However I am having some issues with some basic things - setting macros.

I have created a process to read a DB and discretize - which I am now using as input to several other processes as above. I have set a macro called - path - which I want to propagate to all processes that use this process. However, even though I am setting the properties 'Use Input' and 'Propagate Metadata Recursively' I can't seem to get the path variable into the sub-processes. This is what I really want to do.

However in the interim I also set path variable in the child process using macros in the 'Execute Process' but that does not work either. The 'process_name' macro is not getting activated either. I also tried the 'Set Parameter' operator to do this.

Essentially I need to write out log, result et.al. files to some parametrized paths by process, discretization and such. This has to work from Windows and Linux environments - dev / testing and such.

Please let me know if I am missing some obvious things or maybe need to look up some info.

Also is a P4 3.2 with 4GB of RAM a good enough paltform for development? I am getting some out of memory errors when I run processes with 10 fold XValidation simultaneously  - such as ID3, SVM, Naive Bayes and such that I am experimenting with.

Thanks very much for your help.
Sign In or Register to comment.