Github user mengxr commented on a diff in the pull request:
https://github.com/apache/spark/pull/9749#discussion_r45106138
--- Diff:
mllib/src/test/scala/org/apache/spark/ml/util/DefaultReadWriteTest.scala ---
@@ -64,6 +66,47 @@ trait DefaultReadWriteTest extends TempDirectory { self:
Suite =>
assert(another.uid === instance.uid)
another
}
+
+ /**
+ * Default test for Estimator, Model pairs:
+ * - Explicitly set Params, and train model
+ * - Test save/load using [[testDefaultReadWrite()]] on Estimator and
Model
+ * - Check Params on Estimator and Model
+ * @param estimator Estimator to test
+ * @param dataset Dataset to pass to [[Estimator.fit()]]
+ * @param testParams Set of [[Param]] values to set in estimator
+ * @param checkModelData Method which takes the original and loaded
[[Model]] and compares their
+ * data. This method does not need to check
[[Param]] values.
+ * @tparam E Type of [[Estimator]]
+ * @tparam M Type of [[Model]] produced by estimator
+ */
+ def testEstimatorAndModelReadWrite[E <: Estimator[M] with Writable, M <:
Model[M] with Writable](
+ estimator: E,
+ dataset: DataFrame,
+ testParams: Map[String, Any],
+ checkModelData: (M, M) => Unit): Unit = {
+ // Set some Params to make sure set Params are serialized.
+ testParams.foreach { case (p, v) =>
+ estimator.set(estimator.getParam(p), v)
+ }
+ val model = estimator.fit(dataset)
+
+ // Test Estimator save/load
+ val estimator2 = testDefaultReadWrite(estimator)
+ testParams.foreach { case (p, v) =>
+ val param = estimator.getParam(p)
+ assert(estimator.get(param).get === estimator2.get(param).get)
+ }
+
+ deleteTempDir()
--- End diff --
Shall we modify `testDefaultReadWrite` to append some random string to UID
so calls to it are independent?
---
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
