[GitHub] [spark] zhengruifeng edited a comment on pull request #30013: [SPARK-32455][ML][Follow-Up] LogisticRegressionModel prediction optimization - fix incorrect initialization

[GitBox <gi...@apache.org>]

zhengruifeng edited a comment on pull request #30013:
URL: https://github.com/apache/spark/pull/30013#issuecomment-707430678


   @huaxingao In #29255, if threshold is set in binary lor **Estimator**, then in the **Model**, the values of `_threshold` and `_rawThreshold` are wrong (they are initialized by default `threshold`). If `setThreshold` or `setThresholds` are called, the values of `_threshold` and `_rawThreshold` will be updated to correct values.
   
   > Do we need to somehow improve the test coverage?
   
   Yes, I think we need to figure out why the testsuites failed to catch this, and update them if necessary. 


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org



---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscribe@spark.apache.org
For additional commands, e-mail: reviews-help@spark.apache.org