Github user mengxr commented on the pull request:
https://github.com/apache/spark/pull/1964#issuecomment-54254431
@yu-iskw @erikerlandson @dlwh I prefer simple types for parameters for
model serialization and consistent APIs across languages. In a predictive
model, we should store the training parameters that used to train this model,
and it would be nice to use simple-typed parameters. Another concern is Python
API. If we pass in a distance implementation, we also need to define its Python
counterpart for API consistency, which is not needed by PySpark's k-means
because it calls Scala's implementation through serialization.
For Spark's k-means, it should be good enough to support common and
predefined distance measures, via Breeze.
---
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]
