Well, it finishes, and the data is completely usable, but after I get this:
12/07/17 10:53:30 INFO mapred.Task: Task 'attempt_local_0002_m_000000_0'
done.
12/07/17 10:53:30 INFO mapred.JobClient: map 100% reduce 0%
12/07/17 10:53:30 INFO mapred.JobClient: Job complete: job_local_0002
12/07/17 10:53:30 INFO mapred.JobClient: Counters: 8
12/07/17 10:53:30 INFO mapred.JobClient: File Output Format Counters
12/07/17 10:53:30 INFO mapred.JobClient: Bytes Written=1840447
12/07/17 10:53:30 INFO mapred.JobClient: File Input Format Counters
12/07/17 10:53:30 INFO mapred.JobClient: Bytes Read=3133047
12/07/17 10:53:30 INFO mapred.JobClient: FileSystemCounters
12/07/17 10:53:30 INFO mapred.JobClient: FILE_BYTES_READ=75387890
12/07/17 10:53:30 INFO mapred.JobClient: FILE_BYTES_WRITTEN=75460496
12/07/17 10:53:30 INFO mapred.JobClient: Map-Reduce Framework
12/07/17 10:53:30 INFO mapred.JobClient: Map input records=2771
12/07/17 10:53:30 INFO mapred.JobClient: Spilled Records=0
12/07/17 10:53:30 INFO mapred.JobClient: SPLIT_RAW_BYTES=140
12/07/17 10:53:30 INFO mapred.JobClient: Map output records=2771
12/07/17 10:53:30 INFO driver.MahoutDriver: Program took 121588 ms (Minutes:
2.026466666666667)
It just hangs and I have to manually quit the process. Is this intended
behavior or am I setting some parameter incorrectly or something ? Also, it
appears that the -ow option doesn't work, at least it doesn't work the same
way -ow option works for kmeans
$MAHOUT_HOME/mahout cvb -i ./mahout_data/vectors/vectors/vectors-for-cvb/ -o
./mahout_data/clusters/ -ow -k 80 -dt ./mahout_data/distributions -dict
./mahout_data/vectors/vectors/dictionary.file-0 -mt ./mahout_data/temp/ -x
20 -cd 0.05 -a 10
Thanks,
Seth
--
View this message in context: http://lucene.472066.n3.nabble.com/cvb-doesn-t-finish-tp3995595.html
Sent from the Mahout User List mailing list archive at Nabble.com.
|