Search Mailing List Archives


Limit search to: Subject & Body Subject Author
Sort by: Reverse Sort
Limit to: All This Week Last Week This Month Last Month
Select Date Range     through    

[protege-owl] NCI Thesaurus load problems (NOT MEMORY PROBLEM)

jdmitrie at liacs.nl jdmitrie at liacs.nl
Mon Sep 11 00:09:34 PDT 2006


Hello Tania,

You are asking  me, why I am not converting the Thesaurus.owl in the
database format,
it is because I am not planning to change Thesaurus ( I will also not
working with it from Protege), what I will is only
querying it from my application, and, as I have written long time ago, I am
working with different ontologies and Thesaurus is only the one which
contains
interesting information. The idea is that I use different ontologies in
the same
format.

But ok, if i convert it to database format then I have to work with
OWLDatabaseModel,
is there some tutorial about programming with database model of Protege OWL?

At this moment I load one ontology after another, I get needed information
with
SPARQL, I experimenting with different ontologies and very frequently I have
memory heap space problem. Can it be some problem with Protege memory
management, may be in my case I have to use Jena API? If change all my
ontologies
to database format can I then use reasoner, will OWLDatabaseModel load
everything
to runtime memory, I mean will I still have the memory problems with
database format.

A lot of questions, I am really sorry. I just doubting what I have to use
and what I have
to study more deep.
You every suggestion is very welcome.

Thank you,
Julia

> Hi Julia,
>
> I have suggested changing the repository file with the path of the
> protege.owl ontology, not the import statement of the owl file, which
> should remain like it is. Anyway, the effect is the same.
>
> I don't know why the Protege GUI comes up empty if you give Protege more
> than 1GB of heap. I suppose you get some error in the console.
>
> If you are able to load the thesaurus in Protege OWL, why don't you
> convert it to an OWL database project, because then you wouldn't have
> the heap problem anymore.
>
> Tania
>
>
>
> Julia Dmitrieva wrote:
>
>>Hello Tania,
>>
>>After the changing of Thesaurus.owl file with the
>>good
>><owl:imports
>>rdf:resource="file:///C:/Program%20Files/Protege_3.2_beta/plugins/edu.stanford.smi.protegex.owl/protege.owl"/>
>>I can load it in my java Project.
>>
>>May be have you some suggestions about why Protege cames with empty UI
>>when I try to start it with runtime memory above 1G.
>>
>>Thank you,
>>Julia
>>
>>Julia Dmitrieva wrote:
>>
>>
>>>Thank you, Tania.
>>>
>>>I try  to change the repositories for Thesaurus.owl.
>>>But I have still the Out of memory problem.
>>>I have changed the protege.lax file, and I
>>>have now 2G runtime memory. But if I try to
>>>run Protege with more than 1G memory than
>>>Protege starts with empty screen (I have written
>>>about this problem 2 month ago). Thus I can load
>>>the thesaurus only with 0.9G of memory, but if
>>>I try to do something with it than I have my
>>>lovely problem : Exception in thread "AWT-EventQueue-0"
>>>java.lang.OutOfMemoryError: Java heap spa
>>>ce.
>>>May be can I just change the Thesaurus.owl file with following
>>>In place of:
>>><owl:imports
>>>rdf:resource="http://protege.stanford.edu/plugins/owl/protege"/>
>>>have to be:
>>><owl:imports rdf:resource="file:///C:/Program
>>>Files/Protege_3.2_beta/plugins/edu.stanford.smi.protegex.owl/protege.owl"/>
>>>
>>>In the case of you first solution.
>>>I am using Eclipse and I have added needed jar's to
>>>the class path, I mean Protege.jar and other jar's from
>>>protege/plugins.
>>>
>>>I citate from the The Protégé-OWL API - Programmer's Guide:
>>>"If you are using a Java IDE such as Eclipse or IntelliJ, please select
>>>the Protege installation folder as your project home. Then add all the
>>>jar files from the installation to your project classpath. Set your
>>>compiler output path to plugins/classes."
>>>The probelem is that I am not really understand what it means.
>>>
>>>Do I have to create my project in Protege folder,
>>>thus everething from Protege needs to be loaded?
>>>
>>>
>>>Thank you for help,
>>>Julia
>>>
>>>Tania Tudorache wrote:
>>>
>>>
>>>
>>>>Julia,
>>>>
>>>>The problem is that when you run the program that uses the Protege API,
>>>>it does not find the Protege ontology, which is imported in the NCI
>>>>thesaurus (the ontology is stored as a file "protege.owl" in the owl
>>>>plugins directory). In order to solve this problem, you have two
>>>> solutions:
>>>>
>>>>1. You execute the program in the same directory where Protege is
>>>> installed.
>>>>
>>>>or
>>>>
>>>>2. You edit the repository file of your ontology and add the following
>>>>line to it:
>>>>
>>>>
>>>>file:/C:/MyPath/Protege_3.2_b331/plugins/edu.stanford.smi.protegex.owl/protege.owl?forceReadOnly=true
>>>>
>>>>You replace the path with the path to the local protege.owl file in the
>>>>owl plugins directory. You can do this also from the Protege GUI, in
>>>> the
>>>>"OWL" Menu -> "Ontology repositories" -> "Project repositories" and add
>>>>there a repository for a local file that points to the protege.owl in
>>>>the owl plugins directory.
>>>>
>>>>Tania
>>>>
>>>>
>>>>
>>>>Julia Dmitrieva wrote:
>>>>
>>>>
>>>>
>>>>
>>>>
>>>>>Dear Protege list.
>>>>>
>>>>>I get problems with loading of NCI
>>>>>Thesaurus.owl via Protege API.
>>>>>
>>>>>I can load all triples without problems, but after that
>>>>>i get the following error message.
>>>>>Loaded 950000 triples
>>>>>Loaded 955000 triples
>>>>>[ProtegeOWLParser] Importing
>>>>>http://protege.stanford.edu/plugins/owl/protege (from
>>>>>http://protege.stanford.edu/plugins/owl/protege [actual ontology at
>>>>>this location: null])
>>>>>[ProtegeOWLParser] Completed triple loading after 324485 ms
>>>>>[TripleChangePostProcessor] Completed lists after 156 ms
>>>>>[TripleChangePostProcessor] Completed anonymous classes after 656 ms
>>>>>[TripleChangePostProcessor] Completed deprecated classes after 0 ms
>>>>>[TripleChangePostProcessor] Completed properties after 94 ms
>>>>>
>>>>>:-( :-( :-( :-( And that  :-( :-( :-( :-(
>>>>>
>>>>>java.lang.IllegalArgumentException: null slot
>>>>>  at
>>>>>edu.stanford.smi.protege.model.framestore.ArgumentCheckingFrameStore.assertNotNull(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protege.model.framestore.ArgumentCheckingFrameStore.checkSlot(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protege.model.framestore.ArgumentCheckingFrameStore.getDirectOwnSlotValues(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protege.model.framestore.ModificationFrameStore.getDirectOwnSlotValues(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protege.model.framestore.cleandispatch.DefaultDispatch.getDirectOwnSlotValues(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protege.model.framestore.cleandispatch.CleanDispatchFrameStore.getDirectOwnSlotValues(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protege.model.framestore.FrameStoreAdapter.getDirectOwnSlotValues(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protege.model.framestore.FrameStoreAdapter.getDirectOwnSlotValues(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protegex.owl.model.framestore.OWLFrameStore.getDirectOwnSlotValuesConverting(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protegex.owl.model.impl.OWLUtil.getPropertyValues(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protegex.owl.model.impl.OWLUtil.getPropertyValue(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protegex.owl.model.impl.AbstractRDFSClass.getPropertyValue(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protegex.owl.model.classdisplay.AbstractOWLClassDisplay.getCommentText(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protegex.owl.model.classdisplay.AbstractOWLClassDisplay.getDisplayText(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protegex.owl.model.impl.AbstractOWLAnonymousClass.getBrowserText(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protegex.owl.model.triplestore.impl.DuplicateSuperclassesPostProcessor.removeDuplicateSuperclasses(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protegex.owl.model.triplestore.impl.DuplicateSuperclassesPostProcessor.<init>(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protegex.owl.model.triplestore.impl.TripleChangePostProcessor.postProcess(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protegex.owl.model.triplestore.impl.AbstractTripleStoreModel.endTripleStoreChanges(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protegex.owl.jena.parser.ProtegeOWLParser.loadTriples(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protegex.owl.jena.parser.ProtegeOWLParser.run(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protegex.owl.jena.parser.ProtegeOWLParser.run(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protegex.owl.jena.parser.ProtegeOWLParser.run(Unknown
>>>>>Source)
>>>>>  at edu.stanford.smi.protegex.owl.jena.JenaOWLModel.load(Unknown
>>>>>Source)
>>>>>  at
>>>>>edu.stanford.smi.protegex.owl.ProtegeOWL.createJenaOWLModelFromURI(Unknown
>>>>>Source)
>>>>>  at Thesaurus.thesaurusMining.main(thesaurusMining.java:40)
>>>>>
>>>>>What I find very strange, is that the loading of thesaurus.owl is
>>>>>going OK if I load it from Proptege application.
>>>>>I mean, that it must also be possible to load it via OWL Proptege API.
>>>>>I don't know what I am missing.
>>>>>Do you have some suggestions.
>>>>>
>>>>>Thanks,
>>>>>Julia
>>>>>
>>>>>
>>>>>-------------------------------------------------------------------------
>>>>>To unsubscribe go to
>>>>> http://protege.stanford.edu/community/subscribe.html
>>>>>
>>>>>
>>>>>
>>>>>
>>>>_______________________________________________
>>>>protege-owl mailing list
>>>>protege-owl at lists.stanford.edu
>>>>https://mailman.stanford.edu/mailman/listinfo/protege-owl
>>>>
>>>>
>>>
>>>
>>>
>>
>>
>>
>>
>
> _______________________________________________
> protege-owl mailing list
> protege-owl at lists.stanford.edu
> https://mailman.stanford.edu/mailman/listinfo/protege-owl
>




More information about the protege-owl mailing list