search

Waikato / moa

MOA is an open source framework for Big Data stream mining. It includes a collection of machine learning algorithms (classification, regression, clustering, outlier detection, concept drift detection and recommender systems) and tools for evaluation.
http://moa.cms.waikato.ac.nz/
GNU General Public License v3.0
609 stars 353 forks source link

trying to add new clusterers #186

Closed ElemenT5ven closed 4 years ago

ElemenT5ven commented 4 years ago

Hello,

I'm trying to test if i can add a new clusterers algorithm to MOA but have some issues.

After I have compiled the clusterers with this command in the "lib" directory of the moa release 2019: javac -cp moa.jar MyClusterer.java

It say, in the Moa Manual, to create a moa directory and classifiers directory inside for the creation of new classifiers in manual, so i did the same for clusterers and pasted my MyClusterer.class in a clusterers directory in moa. The moa directory I created is in the lib directory, is it right? Because when i lunch moa.sh in bin directory, I can't find my MyClusterer in the GUI.

What did i do wrong? maybe it's not the right directory to paste my MyClusterer.class? But where?

Thanks,

fracpete commented 4 years ago

Using straight javac is probably not the best way of developing, but it should (in theory) work. Your Java class must use moa.clusterers as package and the class itself must be derived from the abstract superclass moa.clusterers.AbstractClusterer. In the future, please use the appropriate mailing list for your questions: https://moa.cms.waikato.ac.nz/people/contact/