srowen commented on issue #24920: [SPARK-28117][ML] LDA and BisectingKMeans cache the input dataset if necessary URL: https://github.com/apache/spark/pull/24920#issuecomment-505044060 We may need to adjust the test, but I wouldn't disable it. I doubt data is changed, but it may be that caching means the data _doesn't_ change on each iteration. You could simply remove the assertion that there are less than 5 clusters, as I don't see that this is what the test is testing, nor is it invalid to end up with 5 clusters here.
---------------------------------------------------------------- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
