Github user jkbradley commented on a diff in the pull request:
https://github.com/apache/spark/pull/6622#discussion_r31676987
--- Diff:
mllib/src/test/scala/org/apache/spark/ml/classification/OneVsRestSuite.scala ---
@@ -102,6 +103,26 @@ class OneVsRestSuite extends SparkFunSuite with
MLlibTestSparkContext {
val output = ovr.fit(dataset).transform(dataset)
assert(output.schema.fieldNames.toSet === Set("label", "features",
"prediction"))
}
+
+ test("OneVsRest.copy and OneVsRestMode.copy") {
+ val lr = new LogisticRegression()
+ .setMaxIter(1)
+
+ val ovr = new OneVsRest()
+ withClue("copy with classifier unset should work") {
+ ovr.copy(ParamMap(lr.maxIter -> 10))
+ }
+ ovr.setClassifier(lr)
+ val ovr1 = ovr.copy(ParamMap(lr.maxIter -> 10))
+ require(ovr.getClassifier.getOrDefault(lr.maxIter) === 1, "copy should
have no side-effects")
+ require(ovr1.getClassifier.getOrDefault(lr.maxIter) === 10,
+ "copy should handle extra classifier params")
+
+ val ovrModel = ovr1.fit(dataset).copy(ParamMap(lr.threshold -> 0.1))
--- End diff --
Fit using ovr instead of ovr1 (a little faster)
---
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 [email protected] or file a JIRA ticket
with INFRA.
---
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
