[ https://issues.apache.org/jira/browse/SPARK-18334?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15645506#comment-15645506 ]
Yun Ni edited comment on SPARK-18334 at 11/7/16 9:25 PM: --------------------------------------------------------- [~sethah] [~josephkb] Will send a PR to fix shortly. was (Author: yunn): [~sethah][~josephkb] > MinHash should use binary hash distance > --------------------------------------- > > Key: SPARK-18334 > URL: https://issues.apache.org/jira/browse/SPARK-18334 > Project: Spark > Issue Type: Bug > Reporter: Yun Ni > Priority: Trivial > > MinHash currently is using the same `hashDistance` function as > RandomProjection. This does not make sense for MinHash because the Jaccard > distance of two sets is not relevant to the absolute distance of their hash > buckets indices. > This bug could affect accuracy of multi probing NN search for MinHash. -- This message was sent by Atlassian JIRA (v6.3.4#6332) --------------------------------------------------------------------- To unsubscribe, e-mail: issues-unsubscr...@spark.apache.org For additional commands, e-mail: issues-h...@spark.apache.org