mahout-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Suneel Marthi <suneel_mar...@yahoo.com>
Subject Re: MAHOUT 0.9 Release - New URL
Date Sun, 19 Jan 2014 16:30:36 GMT
Its presently setup to run in MR mode (the way its been coded in cluster-reuters.sh). So setting
MAHOUT_LOCAL=true is gonna fail for this.
I am able to see this fail locally when MAHOUT_LOCAL=true.  





On Sunday, January 19, 2014 11:17 AM, Frank Scholten <frank@frankscholten.nl> wrote:
 
Exported MAHOUT_LOCAL=true and still get the same results.



On Sun, Jan 19, 2014 at 5:00 PM, Suneel Marthi <suneel_marthi@yahoo.com>wrote:

> Frank,
>
> Were u running this with MAHOUT_LOCAL=true?
>
>
>
>
>
> On Sunday, January 19, 2014 10:29 AM, Frank Scholten <
> frank@frankscholten.nl> wrote:
>
> -1
>
> The cluster reuters example results in zero clusters when choosing
> streaming k-means. The other steps, unpacking and building do work.
>
> I see this stacktrace:
>
> INFO: Number of Centroids: 0
> Jan 19, 2014 3:51:08 PM org.apache.hadoop.mapred.LocalJobRunner$Job run
> WARNING: job_local797072544_0001
> java.lang.IllegalArgumentException: Must have nonzero number of training
> and test vectors. Asked for %.1f %% of %d vectors for test
> [10.000000149011612, 0]
>     at
> com.google.common.base.Preconditions.checkArgument(Preconditions.java:120)
>     at
> org.apache.mahout.clustering.streaming.cluster.BallKMeans.splitTrainTest(BallKMeans.java:176)
>     at
> org.apache.mahout.clustering.streaming.cluster.BallKMeans.cluster(BallKMeans.java:192)
>     at
> org.apache.mahout.clustering.streaming.mapreduce.StreamingKMeansReducer.getBestCentroids(StreamingKMeansReducer.java:107)
>     at
> org.apache.mahout.clustering.streaming.mapreduce.StreamingKMeansReducer.reduce(StreamingKMeansReducer.java:73)
>     at
> org.apache.mahout.clustering.streaming.mapreduce.StreamingKMeansReducer.reduce(StreamingKMeansReducer.java:37)
>     at org.apache.hadoop.mapreduce.Reducer.run(Reducer.java:177)
>     at
> org.apache.hadoop.mapred.ReduceTask.runNewReducer(ReduceTask.java:649)
>     at org.apache.hadoop.mapred.ReduceTask.run(ReduceTask.java:418)
>     at
> org.apache.hadoop.mapred.LocalJobRunner$Job.run(LocalJobRunner.java:398)
>
> Num clusters: 0; maxDistance: 0.000000
> [Dunn Index] First: Infinity
> [Davies-Bouldin Index] First: NaN
> Jan 19, 2014 3:51:09 PM org.slf4j.impl.JCLLoggerAdapter info
> INFO: Program took 278 ms (Minutes: 0.004633333333333333)
> cluster,distance.mean,distance.sd
> ,distance.q0,distance.q1,distance.q2,distance.q3,distance.q4,count,is.train
>
>
> Here is the full log: http://pastebin.com/TxLV0rDr
>
> As of  yet I am  unfamiliar with the streaming k-means code and the
> algorithms behind it. If anyone has suggestion on what goes wrong in the
> code I am I happy to help  where I can.
>
>
> Frank
>
>
>
> On Sun, Jan 19, 2014 at 10:55 AM, Suneel Marthi <suneel_marthi@yahoo.com>
> wrote:
>
> Thanks Grant.
> >
> >Not sure if I can vote given my role as the BuildMeister/ReleaseMeister
> for 0.9.
> >Here's my +1 FWIW.
> >
> >a) Attached is the draft of the Release notes for 0.9, would definitely
> appreciate feedback on that.
> >
> >b) The vote is open until Monday, Jan 20, 2014 11:59PM EST and passes if
> a majority of atleast 3 +1 PMC votes are cast.
> >
> >The release files, including signatures, digests, etc can be found at:
> >
> https://repository.apache.org/content/repositories/orgapachemahout-1002/org/apache/mahout/mahout-distribution/0.9/
> >
> >The staging repository for this release can be found at:
> >https://repository.apache.org/content/repositories/orgapachemahout-1002
> >
> >Release artifacts have been signed with the following key:
> >https://people.apache.org/keys/committer/smarthi.asc
> >
> >
> >
> >
> >
> >
> >
> >
> >On Saturday, January 18, 2014 12:27 PM, Grant Ingersoll <
> gsingers@apache.org> wrote:
> >
> >Ran the tests, verified sigs, tried out a few of the examples.
> >
> >+1 (binding)
> >
> >
> >On Jan 16, 2014, at 9:41 AM, Suneel Marthi <suneel_marthi@yahoo.com>
> wrote:
> >
> >> Third time's a Charm!!!
> >>
> >>
> >> Here's the new URL for Mahout 0.9 Release:
> >>
> https://repository.apache.org/content/repositories/orgapachemahout-1002/org/apache/mahout/mahout-distribution/0.9/
> >>
> >> For those volunteering to test this, some of the things to be verified:
> >>
> >> a) Verify that u can
>  unpack the release (tar or zip)
> >> b) Verify u r able to compile the distro
> >> c)  Run through the unit tests: mvn clean test
> >> d) Run the example scripts under $MAHOUT_HOME/examples/bin. Please run
> through all the different options in each script.
> >>
> >>
> >> Committers
> >> and PMC members:
> >> ---------------------------------------
> >>
> >> Need 'at least 3 +1 votes' for the Release to pass.
> >>
> >>
> >> Thanks and Regards.
> >
> >
> >
> >
>
Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message