[
https://issues.apache.org/jira/browse/MAHOUT-30?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12659282#action_12659282
]
Isabel Drost commented on MAHOUT-30:
------------------------------------
> Regarding the question of whether something should be called a Distribution
> or a Sampler, the mathematical terminology is that a distribution is
> something you can sample so the the Distribution terminology would be most
> compatible that way. The fact that only one method is currently
> defined is likely a temporary thing ... other methods could well be required
> for later efforts.
I understand. I do not have any strong objections. I think, a short class
comment in DirichletDistribution would already help to avoid at least my
confusion. (Although not trying to understand code at 2a.m. local time might
help as well... ;) )
> I was thinking of moving the display code into the examples directory.
Sounds like a great idea to me.
> I did that so Ted could use his favorite library but he has not been pursuing
> it. I'm happy with blog and, as commons does not have the
> needed sampling methods without Ted's patches, suggest we could go with blog.
> Removing the plugability would clean up the code some too.
Do you know what the current status of the patches is? I must admit I have a
slight preference for commons-math as well, in case they support what we need.
> dirichlet process implementation
> --------------------------------
>
> Key: MAHOUT-30
> URL: https://issues.apache.org/jira/browse/MAHOUT-30
> Project: Mahout
> Issue Type: New Feature
> Components: Clustering
> Reporter: Isabel Drost
> Assignee: Jeff Eastman
> Attachments: jeastman.vcf, MAHOUT-30.patch, MAHOUT-30b.patch,
> MAHOUT-30c.patch
>
>
> Copied over from original issue:
> > Further extension can also be made by assuming an infinite mixture model.
> > The implementation is only slightly more difficult and the result is a
> > (nearly)
> > non-parametric clustering algorithm.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
