[ https://issues.apache.org/jira/browse/MAHOUT-99?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12683376#action_12683376 ]
Richard Tomsett commented on MAHOUT-99: --------------------------------------- I tried just reverting back to SequenceFiles in the test and KMeansUtil classes but couldn't get the test to complete correctly - must admit I didn't work on it too long as it was late... I'm not sure quite what the problem was, but I think it was just that I hadn't found all the relevant changes that needed to be made. I guess if you've found it in the Canopy class (I take it that was modified in this patch NOT to output SequenceFiles as well) that would explain it :) > Improving speed of KMeans > ------------------------- > > Key: MAHOUT-99 > URL: https://issues.apache.org/jira/browse/MAHOUT-99 > Project: Mahout > Issue Type: Improvement > Components: Clustering > Reporter: Pallavi Palleti > Assignee: Grant Ingersoll > Fix For: 0.1 > > Attachments: MAHOUT-99-1.patch, Mahout-99.patch, MAHOUT-99.patch > > > Improved the speed of KMeans by passing only cluster ID from mapper to > reducer. Previously, whole Cluster Info as formatted s`tring was being sent. > Also removed the implicit assumption of Combiner runs only once approach and > the code is modified accordingly so that it won't create a bug when combiner > runs zero or more than once. -- This message is automatically generated by JIRA. - You can reply to this email to add a comment to the issue online.