Github user erikerlandson commented on a diff in the pull request:
https://github.com/apache/spark/pull/13440#discussion_r104822183
--- Diff:
mllib/src/main/scala/org/apache/spark/mllib/tree/impurity/Impurity.scala ---
@@ -50,6 +50,50 @@ trait Impurity extends Serializable {
@Since("1.0.0")
@DeveloperApi
def calculate(count: Double, sum: Double, sumSquares: Double): Double
+
+ /**
+ * :: DeveloperApi ::
+ * Compute a test-statistic p-value quality measure from left and right
split populations
+ * @param calcL impurity calculator for the left split population
+ * @param calcR impurity calculator for the right split population
+ * @return The p-value for the null hypothesis; that left and right
split populations
+ * represent the same distribution
+ * @note Unless overridden this method will fail with an exception, for
backward compatability
+ */
+ @Since("2.0.0")
+ @DeveloperApi
+ def calculate(calcL: ImpurityCalculator, calcR: ImpurityCalculator):
Double =
+ throw new UnsupportedOperationException("Impurity.calculate")
+
+ /**
+ * :: DeveloperApi ::
+ * Determine if this impurity measure is a test-statistic measure
+ * @return True if this is a split quality measure based on a test
statistic (i.e. returns a
+ * p-value) or false otherwise.
+ * @note Unless overridden this method returns false by default, for
backward compatability
+ */
+ @Since("2.0.0")
+ @DeveloperApi
+ def isTestStatistic: Boolean = false
+}
+
+/**
+ * :: DeveloperApi ::
+ * Utility functions for Impurity measures
+ */
+@Since("2.0.0")
+@DeveloperApi
--- End diff --
I don't think so. I don't recall any specific motivation to keep it
private, but historically Spark seems to default things to "minimum
visibility." The only method currently defined here is an implementation
detail for hacking p-values into the existing 'gain' system, where larger is
assumed to be better.
---
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 infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
