Github user filousen commented on the issue: https://github.com/apache/spark/pull/16355 @imatiach-msft thank you for checking this. I'm using a VectorAssembler to transform the dataset I read from the json I pasted earlier. VectorAssembler assembler = new VectorAssembler() .setInputCols(new String[]{"count"}) .setOutputCol("features"); Dataset<Row> ds = assembler.transform(jsonDs). Then I use the transformed dataset with BisectingKMeans. BisectingKMeans bkm = new BisectingKMeans().setK(100).setFeaturesCol("features").setMaxIter(2); BisectingKMeansModel model = bkm.fit(ds); Dataset<Row> res = model.transform(ds); I have had about 30% of my datasets that failed with the similar error. Thanks
--- 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