You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@mahout.apache.org by "Suneel Marthi (JIRA)" <ji...@apache.org> on 2014/01/22 09:58:20 UTC
[jira] [Comment Edited] (MAHOUT-1402) Zero clusters using streaming
k-means option in cluster-reuters.sh
[ https://issues.apache.org/jira/browse/MAHOUT-1402?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13878304#comment-13878304 ]
Suneel Marthi edited comment on MAHOUT-1402 at 1/22/14 8:57 AM:
----------------------------------------------------------------
The MR version of Streaming KMeans seems to be failing (the sequential mode passes), the reason being that the reducer is reading zero intermediate centroids; need to investigate as to what's going on.
was (Author: smarthi):
The MR version of Streaming KMeans seems to be failing (the sequential mode passes), the reason being that the reducer is reading zero centroids from the mappers; need to investigate as to what's going on.
> Zero clusters using streaming k-means option in cluster-reuters.sh
> ------------------------------------------------------------------
>
> Key: MAHOUT-1402
> URL: https://issues.apache.org/jira/browse/MAHOUT-1402
> Project: Mahout
> Issue Type: Bug
> Components: Clustering
> Affects Versions: 0.8
> Environment: AWS default Linux AMI
> Reporter: Andrew Musselman
> Assignee: Suneel Marthi
> Fix For: 0.9
>
>
> Running cluster-reuters.sh in examples/bin results in this:
> [snip]
> INFO: Number of Centroids: 0
> Jan 22, 2014 1:52:22 AM org.apache.hadoop.mapred.LocalJobRunner$Job run
> WARNING: job_local23982482_0001
> java.lang.IllegalArgumentException: Must have nonzero number of training and test vectors. Asked for %.1f %% of %d vectors for test [10.000000149011612, 0]
> at com.google.common.base.Preconditions.checkArgument(Preconditions.java:120)
> at org.apache.mahout.clustering.streaming.cluster.BallKMeans.splitTrainTest(BallKMeans.java:176)
> at org.apache.mahout.clustering.streaming.cluster.BallKMeans.cluster(BallKMeans.java:192)
> at org.apache.mahout.clustering.streaming.mapreduce.StreamingKMeansReducer.getBestCentroids(StreamingKMeansReducer.java:107)
> at org.apache.mahout.clustering.streaming.mapreduce.StreamingKMeansReducer.reduce(StreamingKMeansReducer.java:73)
> at org.apache.mahout.clustering.streaming.mapreduce.StreamingKMeansReducer.reduce(StreamingKMeansReducer.java:37)
> at org.apache.hadoop.mapreduce.Reducer.run(Reducer.java:177)
> at org.apache.hadoop.mapred.ReduceTask.runNewReducer(ReduceTask.java:649)
> at org.apache.hadoop.mapred.ReduceTask.run(ReduceTask.java:418)
> at org.apache.hadoop.mapred.LocalJobRunner$Job.run(LocalJobRunner.java:398)
> [snip]
> WARNING: No qualcluster.props found on classpath, will use command-line arguments only
> Num clusters: 0; maxDistance: 0.000000
> [Dunn Index] First: Infinity
> [Davies-Bouldin Index] First: NaN
> Jan 22, 2014 1:52:24 AM org.slf4j.impl.JCLLoggerAdapter info
> INFO: Program took 535 ms (Minutes: 0.008916666666666666)
> cluster,distance.mean,distance.sd,distance.q0,distance.q1,distance.q2,distance.q3,distance.q4,count,is.train
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)