[
https://issues.apache.org/jira/browse/MAHOUT-479?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12899710#action_12899710
]
Jeff Eastman commented on MAHOUT-479:
-------------------------------------
I'm making some progress on integrating the clustering data structures. Here's
how the top levels are shaping up:
{code}
// this defines the minimal Dirichlet model
public interface Model<O> extends Writable {
double pdf(O x);
void observe(O x);
int count();
void computeParameters();
public Model<VectorWritable> sampleFromPosterior();
}
// this puts a face on a cluster
public interface Cluster {
int getId();
Vector getCenter();
Vector getRadius();
int getNumPoints();
String asFormatString(String[] bindings);
String asJsonString();
}
// here's the resulting Cluster class hierarchy
public abstract class AbstractCluster implements Cluster, Model<VectorWritable>
{}
public abstract class DistanceMeasureCluster extends AbstractCluster {}
public class Canopy extends DistanceMeasureCluster {}
public class Cluster extends DistanceMeasureCluster {}
public class MeanShiftCanopy extends Cluster {}
public class SoftCluster extendsCluster {}
public class GaussianCluster extends AbstractCluster {}
// Note: all the current Dirichlet models can be subsumed by GaussianCluster or
simple subclasses
{code}
There's a fair amount of cleanup to tests etc to do but I thought I'd post this
for visibility.
> Streamline classification/ clustering data structures
> -----------------------------------------------------
>
> Key: MAHOUT-479
> URL: https://issues.apache.org/jira/browse/MAHOUT-479
> Project: Mahout
> Issue Type: Improvement
> Components: Classification, Clustering
> Affects Versions: 0.1, 0.2, 0.3, 0.4
> Reporter: Isabel Drost
>
> Opening this JIRA issue to collect ideas on how to streamline our
> classification and clustering algorithms to make integration for users easier
> as per mailing list thread http://markmail.org/message/pnzvrqpv5226twfs
> {quote}
> Jake and Robin and I were talking the other evening and a common lament was
> that our classification (and clustering) stuff was all over the map in terms
> of data structures. Driving that to rest and getting those comments even
> vaguely as plug and play as our much more advanced recommendation components
> would be very, very helpful.
> {quote}
> This issue probably also realates to MAHOUT-287 (intention there is to make
> naive bayes run on vectors as input).
> Ted, Jake, Robin: Would be great if someone of you could add a comment on
> some of the issues you discussed "the other evening" and (if applicable) any
> minor or major changes you think could help solve this issue.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
