[Central] Prom 6.0 Tutorial questions
Hello everyone,
I noticed there is no topic with questions regarding tot the Prom 6.0 tutorial. Therefore I think it is a good idea to create one where all questions regarding the Prom tutorial can be asked and answered.
Here is mine:
I am currently going through the Prom 6.0 tutorial and one thing that already stands out is the time it takes (4 hours and counting) to clean a log as described in chapter 2.2. I was unable to find 'simple log filter' so I used 'Filter Log using Simple Heuristics' which appears to do the same thing. I am hoping that there is a way to speed up the process. In another topic I read you can increase the allocation of memory as described in this post: http://www.win.tue.nl/promforum/discussion/108/performance-with-large-logs#Item_2.
This is what is currently written in my .bat file:
@setlocal enableextensions@cd /d "%~dp0"jre7\bin\java -da -Xmx1G -XX:MaxPermSize=256m -classpath ProM65.jar -Djava.util.Arrays.useLegacyMergeSort=true org.processmining.contexts.uitopia.UI
In the post Joos prescribes to increase -Xmx1g to a higher number. My computer has 8 gigabytes of working memory but the computers at the company where I am writing my thesis have only 4 gigabytes of working memory. What would be sensible numbers to assign to these computers? Moreover, should the company look to invest in extra working memory to increase the level of analysis (currently i3 processor with 4g ram)?
Edit: It says: java.nio.BufferUnderflowException while executing
Best Answers
-
Dear Erik,
Great idea to start this thread.
Regarding your first question, filtering an event log for 4 hours is way too long. This should be done in minutes, worst case. One thing you could try to do is to install the XESLite package. Then, when you import an event log, you get to choose how memory (in)efficient the event log is loaded. However, some import methods don't allow you to edit the event log (e.g. filtering).
Your long runtime is likely to be caused by the exception that you mention. Could you run ProM from the .bat file or command line? If the error occurs again, could you copy the whole message, including stacktrace, here? That would help pinpointing the cause.
Joos Buijs
Senior Data Scientist and process mining expert at APG (Dutch pension fund executor).
Previously Assistant Professor in Process Mining at Eindhoven University of Technology -
Dear Erik,
I was made aware of this thread as someone in our group encountered the same issue with the "Filter Log using Simple Heuristics" plug-in. Unfortunately, an error slipped into the "Sequential" option of XESLite in the ProM 6.5 release. So please use the standard option "Disk Buffered by MapDB" if you want to use MapDB as backend.
The "Sequential" option is actually experimental, but the current import plug-in architecture is very limited, and does not allow me to give any hint or help text on the maturity of importer plug-ins.
I'm working on a bugfix for the next release ...
Answers
-
Dear Erik,
Thanks for this thread. It also looks like the 'alpha-algorithm' is the 'Mine for a Petri-Net using the Alpha-algorithm' action.
I allocated 4Gb on an 8Gb machine as per above. The repairexample processes in seconds.
Regards
Peter
Howdy, Stranger!
Categories
- 1.6K All Categories
- 45 Announcements / News
- 225 Process Mining
- 6 - BPI Challenge 2020
- 9 - BPI Challenge 2019
- 24 - BPI Challenge 2018
- 27 - BPI Challenge 2017
- 8 - BPI Challenge 2016
- 68 Research
- 1K ProM 6
- 394 - Usage
- 288 - Development
- 9 RapidProM
- 1 - Usage
- 7 - Development
- 54 ProM5
- 19 - Usage
- 187 Event Logs
- 32 - ProMimport
- 75 - XESame