bscan commented on a change in pull request #23705: [SPARK-26787] Fix
standardizeLabels error message in WeightedLeastSquares
URL: https://github.com/apache/spark/pull/23705#discussion_r252782228
##########
File path:
mllib/src/main/scala/org/apache/spark/ml/optim/WeightedLeastSquares.scala
##########
@@ -133,7 +133,8 @@ private[ml] class WeightedLeastSquares(
return new WeightedLeastSquaresModel(coefficients, intercept,
diagInvAtWA, Array(0D))
} else {
require(!(regParam > 0.0 && standardizeLabel), "The standard deviation
of the label is " +
- "zero. Model cannot be regularized with standardization=true")
+ "zero. Model cannot be regularized with standardizeLabel=true " +
+ "(standardizeLabel is not exposed in the LinearRegression API)")
Review comment:
Good point. Since the primary way to get here is through calling
LinearRegression(), I was trying to give the user a message that indicates it's
a current limitation of the API, not the implementation/algorithm. I updated to
your suggestion though as I do think it's an improvement over the current
message of `standardization=true`, which was simply incorrect. Thanks for
reviewing!
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
