[
https://issues.apache.org/jira/browse/SPARK-6398?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14623036#comment-14623036
]
Feynman Liang commented on SPARK-6398:
--------------------------------------
Possibly redundant with SPARK-7210? Dimensionality's primary contribution to
the method's utility is the matrix inversion while calculating the pdf in the
M-step, so a better inversion method will solve this issue as well.
> Improve utility of GaussianMixture for higher dimensional data
> --------------------------------------------------------------
>
> Key: SPARK-6398
> URL: https://issues.apache.org/jira/browse/SPARK-6398
> Project: Spark
> Issue Type: Improvement
> Components: MLlib
> Reporter: Travis Galoppo
> Assignee: Travis Galoppo
>
> The current EM implementation for GaussianMixture protects itself from
> numerical instability at the expense of utility in high dimensions. A few
> options exist for extending utility into higher dimensions.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
