[
https://issues.apache.org/jira/browse/FLINK-940?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14035715#comment-14035715 ]
Bastian Köcher commented on FLINK-940:
--------------------------------------
Okay I played a little bit more with the output, now I found something which could describe the problem.
"value = (4.0, 49, 1691)
Merging document (1669, 49) into cluster 1691"
This is the output I'm getting, the first line is the smallest similarity found in this round, consisting of the (similarity, clusterid0, clusterid1) and the second line is the output of the merge which is done, consisting of the (clusterid, documentid).
"DataSet<Document> delta = linkageSet.coGroup(iteration.getSolutionSet()).where(1).equalTo(0).with(new PointUpdaterReverse());"
I want to compare the key at position 1 of the linkageSet (clusterid0) with the solutionset key at position 0(clusterid).
But the output states that it compares the key at position 1 of the linkageSet with the solutionset key at position 1(documentid).
> DeltaIteration coGroup at SolutionSet doesn't find values
> ---------------------------------------------------------
>
> Key: FLINK-940
> URL:
https://issues.apache.org/jira/browse/FLINK-940> Project: Flink
> Issue Type: Bug
> Components: Java API
> Affects Versions: pre-apache-0.5.1
> Reporter: Bastian Köcher
> Assignee: Stephan Ewen
> Attachments: IMPRO-3.SS14.G03.zip
>
>
> If I execute a coGroup on a SolutionSet, the grouping doesn't find any matching tuples, but if I reverse the grouping(SolutionSet as second set), matching tuples are found.
--
This message was sent by Atlassian JIRA
(v6.2#6252)
Locked
