Hi, This error seems to be caused by the ClusterDumper in mahout 0.6. With the same input files, if I use the ClusterDumper in mahout 0.5, everything works fine.
Huanchen On Apr 17, 2012, at 1:24 AM, Paritosh Ranjan wrote: > Its very difficult to think of the problem with only this much info. > Can you try out few things: > > 1) Try out the same clustering with 0.7-snapshot. > 2) Try out both standalone and cluster mode ( but unsetting/setting > HADOOP_HOME ), for both 0.6 and 0.7-snapshot. Then at least we will know were > the problem is. > > On 17-04-2012 10:47, Huanchen Zhang wrote: >> I'm trying k-means. Thank you for your reply. >> >> On Apr 16, 2012, at 9:19 PM, Paritosh Ranjan wrote: >> >>> Which clustering algorithm did you select for execution in cluster-reuters? >>> >>> On 17-04-2012 08:08, Huanchen Zhang wrote: >>>> hi, >>>> >>>> I'm running mahout cluster-reuters example on amazon ec2 from my local >>>> client. I'm using mahout 0.6, hadoop 0.20.2. >>>> I always get the same following errors: >>>> >>>> 12/04/16 19:32:45 INFO mapred.JobClient: Task Id : >>>> attempt_201204170143_0029_m_000000_0, Status : FAILED >>>> java.lang.NullPointerException >>>> at >>>> org.apache.mahout.clustering.evaluation.RepresentativePointsMapper.mapPoint(RepresentativePointsMapper.java:73) >>>> at >>>> org.apache.mahout.clustering.evaluation.RepresentativePointsMapper.map(RepresentativePointsMapper.java:60) >>>> at >>>> org.apache.mahout.clustering.evaluation.RepresentativePointsMapper.map(RepresentativePointsMapper.java:40) >>>> at org.apache.hadoop.mapreduce.Mapper.run(Mapper.java:144) >>>> at org.apache.hadoop.mapred.MapTask.runNewMapper(MapTask.java:621) >>>> at org.apache.hadoop.mapred.MapTask.run(MapTask.java:305) >>>> at org.apache.hadoop.mapred.Child.main(Child.java:170) >>>> >>>> 12/04/16 19:32:45 WARN mapred.JobClient: Error reading task >>>> outputip-10-195-30-47.ec2.internal >>>> 12/04/16 19:32:45 WARN mapred.JobClient: Error reading task >>>> outputip-10-195-30-47.ec2.internal >>>> >>>> Anyone has any idea? Thank you very much! >>>> >>>> Best, >>>> Huanchen >>>> >>>> >
