[ https://issues.apache.org/jira/browse/SPARK-31676?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Apache Spark reassigned SPARK-31676: ------------------------------------ Assignee: Apache Spark > QuantileDiscretizer raise error parameter splits given invalid value (splits > array includes -0.0 and 0.0) > --------------------------------------------------------------------------------------------------------- > > Key: SPARK-31676 > URL: https://issues.apache.org/jira/browse/SPARK-31676 > Project: Spark > Issue Type: Bug > Components: ML > Affects Versions: 2.4.5, 3.0.0 > Reporter: Weichen Xu > Assignee: Apache Spark > Priority: Major > > Reproduce code > {code} > import scala.util.Random > val rng = new Random(3) > val a1 = Array.tabulate(200)(_=>rng.nextDouble * 2.0 - 1.0) ++ > Array.fill(20)(0.0) ++ Array.fill(20)(-0.0) > import spark.implicits._ > val df1 = sc.parallelize(a1, 2).toDF("id") > import org.apache.spark.ml.feature.QuantileDiscretizer > val qd = new > QuantileDiscretizer().setInputCol("id").setOutputCol("out").setNumBuckets(200).setRelativeError(0.0) > val model = qd.fit(df1) > {code} > Raise error like: > at org.apache.spark.ml.param.Param.validate(params.scala:76) > at org.apache.spark.ml.param.ParamPair.<init>(params.scala:634) > at org.apache.spark.ml.param.Param.$minus$greater(params.scala:85) > at org.apache.spark.ml.param.Params.set(params.scala:713) > at org.apache.spark.ml.param.Params.set$(params.scala:712) > at org.apache.spark.ml.PipelineStage.set(Pipeline.scala:41) > at org.apache.spark.ml.feature.Bucketizer.setSplits(Bucketizer.scala:77) > at > org.apache.spark.ml.feature.QuantileDiscretizer.fit(QuantileDiscretizer.scala:231) > ... 49 elided > java.lang.IllegalArgumentException: quantileDiscretizer_479bb5a3ca99 > parameter splits given invalid value [-Infinity,-0.9986765732730827,..., > -0.0, 0.0, ..., 0.9907184077958491,Infinity] > 0.0 > -0.0 is False, which break the paremater validation check. -- This message was sent by Atlassian Jira (v8.3.4#803005) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org