Re: IB1 Classifier
Peter Reutemann <[email protected]>
| Newsgroups | gmane.comp.ai.weka |
|---|---|
| Message-ID | <[email protected]> |
If you're using maven or gradle as build system, just pull the artifacts from maven central. See snippets here: https://search.maven.org/artifact/nz.ac.waikato.cms.weka/simpleEducationalLearningSchemes/1.0.2/jar If you include them manually in your projects, you can download the jar(s) from maven central (see above link). If you want to keep using the weka package manager, then include the following in your code to load the packages: weka.gui.GenericObjectEditor.determineClasses(); You can then instantiate classes from packages via the forName method of the weka.core.Utils class. Cheers, Peter On August 19, 2021 11:26:11 PM GMT+12:00, Joseph McVerry <[email protected]> wrote: > The weka files are stored in a directory using the version numbers. > >By choice, I don't use a CLASSPATH environment variable. I set the classpath at runtime to avoid conflicts. > >So my real problem is accessing the Package Manager updated class files from a Java program. What class directory or jar file needs to be added to the classpath to access files added using the Package Manager? > > >Joe McVerry > >http://mcverryreport.com - Links to Stock Trader Blogs and Reports > > > > On Thursday, August 19, 2021, 02:34:43 AM EDT, Peter Reutemann <[email protected]> wrote: > > Rename the "wekafiles" directory (in your home directory) before starting up Weka. That ensures that you don't have any access to previously installed packages. > >Also, make sure that you don't have any old Weka jars on your CLASSPATH environment variable. > >Now start up Weka an check again. > >Cheers, Peter > >On August 19, 2021 6:22:11 PM GMT+12:00, Joseph McVerry <[email protected]> wrote: >The IB1 classifier is available in my version (3.9.5) of the Weka Explorer, even though I never installed it using the Package Manager. >When I upgrade my version of Weka does the upgrader look to see what I have installed? > >Joe McVerry > >http://mcverryreport.com - Links to Stock Trader Blogs and Reports > > On Wednesday, August 18, 2021, 09:46:33 PM EDT, Peter Reutemann <[email protected]> wrote: > > > Presently I am using 3.9.5. >> >> I ran the jar.exe using the tf option against the weka.jar file. Here is a list of the lazy classifiers: >> >> ... >> weka/classifiers/lazy/ >> weka/classifiers/lazy/kstar/ >> ... >> weka/classifiers/lazy/IBk.class >> weka/classifiers/lazy/KStar.class >> weka/classifiers/lazy/LWL.class >> weka/classifiers/lazy/kstar/KStarCache$CacheTable.class >> weka/classifiers/lazy/kstar/KStarCache$TableEntry.class >> weka/classifiers/lazy/kstar/KStarCache.class >> weka/classifiers/lazy/kstar/KStarConstants.class >> weka/classifiers/lazy/kstar/KStarNominalAttribute.class >> weka/classifiers/lazy/kstar/KStarNumericAttribute.class >> weka/classifiers/lazy/kstar/KStarWrapper.class >> ... >> >> >> As you can see no IB1.class file is in the jar file. > >Which aligns with what I wrote earlier: > >Only older versions (pre 3.8, e.g., 3.6.15) will have IB1 as part of >the weka.jar. Newer versions, as Eibe pointed out, make it available >via a package. IBk is still available in the base weka.jar in newer >versions of Weka. > >Below is the Weka package Eibe referenced that you need to install in >order to get the IB1 class (if you don't want to use IBk with k=1): >https://weka.sourceforge.io/packageMetaData/simpleEducationalLearningSchemes/index.html > >Cheers, Peter >-- >Peter Reutemann >Dept. of Computer Science >University of Waikato, NZ >+64 (7) 858-5174 (office) >+64 (7) 577-5304 (home office) >http://www.cms.waikato.ac.nz/~fracpete/ >http://www.data-mining.co.nz/._______________________________________________ >Wekalist mailing list -- [email protected] >Send posts to [email protected] >To unsubscribe send an email to [email protected] >To subscribe, unsubscribe, etc., visit https://list.waikato.ac.nz/postorius/lists/wekalist.list.waikato.ac.nz >List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html > -- Peter Reutemann Dept. of Computer Science University of Waikato, NZ +64 (7) 858-5174 (office) +64 (7) 577-5304 (home office) http://www.cms.waikato.ac.nz/~fracpete/ http://www.data-mining.co.nz/. _______________________________________________ Wekalist mailing list -- [email protected] Send posts to [email protected] To unsubscribe send an email to [email protected] To subscribe, unsubscribe, etc., visit https://list.waikato.ac.nz/postorius/lists/wekalist.list.waikato.ac.nz List etiquette: http://www.cs.waikato.ac.nz/~ml/weka/mailinglist_etiquette.html