The RapidMiner community is on read-only mode until further notice. Technical support via cases will continue to work as is. For any urgent licensing related requests from Students/Faculty members, please use the Altair academic forum here.
CrossValidation logging performance
marcin_blachnik
Member Posts: 61 
Guru
in Help
Hi
I have noticed that the performance which is logged by the Log operator from the CrossValidation operator is actually microaverage while the standard deviation is logged from the regular (macro) performance, so these two values (Performance + Standard deviation do not match - these are two independent values. To log the actual average performance and standard deviation it is necessary to transform performance vector to data and log values from the data or use some other tricks.
If it's possible I'd suggest adding the possibility to log both micro and macro performance from CrossValidation, so that the average and standard deviation match.
Best
Marcin
I have noticed that the performance which is logged by the Log operator from the CrossValidation operator is actually microaverage while the standard deviation is logged from the regular (macro) performance, so these two values (Performance + Standard deviation do not match - these are two independent values. To log the actual average performance and standard deviation it is necessary to transform performance vector to data and log values from the data or use some other tricks.
If it's possible I'd suggest adding the possibility to log both micro and macro performance from CrossValidation, so that the average and standard deviation match.
Best
Marcin
Tagged:
0
