[
https://issues.apache.org/jira/browse/SPARK-11834?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15260328#comment-15260328
]
Jeremy commented on SPARK-11834:
--------------------------------
The upside to this change is that users who set a threshold won't silently
clear it by also calling thresholds. Users should never call this anyways, as
they'll only be doing binary classification - it's unlikely that anybody has
thresholds in their current usage pattern. The main downside that I see is that
it would need to be changed back after multi is added. If that event is far
away this change may save a few users some confusion, but if it's imminent then
it would be work to undo this change quite soon. Regardless, the change is only
consequential in a pretty isolated case that involves using the function
incorrectly - quite minor either way.
> Ignore thresholds in LogisticRegression and update documentation
> ----------------------------------------------------------------
>
> Key: SPARK-11834
> URL: https://issues.apache.org/jira/browse/SPARK-11834
> Project: Spark
> Issue Type: Improvement
> Components: Documentation, ML
> Affects Versions: 1.6.0
> Reporter: Xiangrui Meng
> Assignee: Xiangrui Meng
> Priority: Minor
>
> ml.LogisticRegression does not support multiclass yet. So we should ignore
> `thresholds` and update the documentation. In the next release, we can do
> SPARK-11543.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
