Github user mengxr commented on the pull request:
https://github.com/apache/spark/pull/3519#issuecomment-68286932
2a) `(label: Double, feature: Double, weight: Double)` sounds good to me.
We may add weight support to `LabeledPoint` as part of SPARK-3702, which should
be orthogonal to this PR. We can update the API here (before 1.3) once that
gets merged.
2b) Isotonic regression is a univariate regression algorithm. It is not
necessary to have its model extend RegressionModel. It should have
`predict(RDD[Double])` and `predict(Double)`.
2c) Try `train(JavaPairRDD<java.lang.Double, java.lang.Double>)`
---
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]
