Github user jkbradley commented on the pull request:
https://github.com/apache/spark/pull/3022#issuecomment-67422512
Thanks for the updates! I've started running some quick tests, and it's
made me think of some more items. I've added some inline comments as well; the
main issue Iâve run into is the failure for covariance matrices which are not
full rank.
The new prediction methods look useful. How would you feel about this set
of methods:
* predict(): predict best cluster as an Int for each data point (same as in
KMeans)
* predictMembership(): predict membership in all clusters as a Vector for
each data point (predictClusters is more ambiguous, IMO.)
Also, can the prediction code be moved to within the model class?
---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
