zhengruifeng commented on a change in pull request #26135:
[SPARK-29489][ML][PySpark] ml.evaluation support log-loss
URL: https://github.com/apache/spark/pull/26135#discussion_r335914224
##########
File path: python/pyspark/ml/evaluation.py
##########
@@ -365,18 +375,25 @@ class MulticlassClassificationEvaluator(JavaEvaluator,
HasLabelCol, HasPredictio
"The beta value used in weightedFMeasure|fMeasureByLabel."
" Must be > 0. The default value is 1.",
typeConverter=TypeConverters.toFloat)
+ eps = Param(Params._dummy(), "eps",
Review comment:
SKLearn has this param (default value is 1e-15), I follow sklearn's design.
Other lib like xgboost also use an epsilon to compute
[log-loss](https://github.com/dmlc/xgboost/blob/master/src/objective/regression_loss.h#L72),
but it's not exposed to end user and its default value is 1e-16.
----------------------------------------------------------------
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:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
