Github user srowen commented on a diff in the pull request:
https://github.com/apache/spark/pull/6684#discussion_r31868395
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/regression/StreamingLinearAlgorithm.scala
---
@@ -79,9 +79,6 @@ abstract class StreamingLinearAlgorithm[
* @param data DStream containing labeled data
*/
def trainOn(data: DStream[LabeledPoint]): Unit = {
- if (model.isEmpty) {
--- End diff --
No, I think the user must set the weights before starting. Above it says
"Initial weights must be set before using trainOn or predictOn". I am not sure
it makes sense for the scaladoc to talk about a default, therefore. Yes
`Vectors.dense(numFeatures)` is a bug but I suppose it is never hit since the
model is always defined at this point. I think this can be changed along with
the comment. Yeah, that is probably worth touching up.
---
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]
