mahout-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Lance Norskog <goks...@gmail.com>
Subject Re: Error in executing mahout kmeans
Date Tue, 22 Nov 2011 03:20:31 GMT
Dipesh-

To run the Reuters dataset, use examples/bin/build-reuters.sh. There are a
lot of options and it is easier to see how it all works.

DisplayKMeans is a standalone Swing program that shows a small fabricated
set of points as an educational tool. It does not show your data.

If you want to do that, there is an option to export clusters into a
displayable format called 'graphml'. When you have your clusters created,
run 'mahout clusterdump'. Use 'output format' of GML. There is a separate
app called 'Gephi' that can read files in this format.

On Mon, Nov 21, 2011 at 7:10 PM, DIPESH KUMAR SINGH
<dipesh.tech@gmail.com>wrote:

> Mahout is installed as i can get some output on executing $mahout
>
> I am not sure all the libraries are build or not.
>
> Just want to know presence of which all files would let me know that mahout
> is build? (or any folder etc)
>
>
>
>
>
> On Tue, Nov 22, 2011 at 1:16 AM, Dan Beaulieu
> <danjacob.beaulieu@gmail.com>wrote:
>
> > Have you built mahout? You'll need to do that via:
> > $mvn install
> >
> >
> >
> > On Monday, November 21, 2011, DIPESH KUMAR SINGH <dipesh.tech@gmail.com>
> > wrote:
> > > I am unable to figure out how to use .job file.
> > >
> > > Do i need to build the DisplayKmeans.java file, by compiling (javac)
> and
> > > making jars etc.?
> > >
> > > To get started, i was trying to just run kmeans example in mahout from
> > CLI.
> > >
> > > I could make the hadoop sequence files in hdfs, but on running
> > seq2sparse,
> > > i am getting following 2 errors.
> > >
> > > (I was following this ppt :
> > > Link<
> >
> >
> http://assets.en.oreilly.com/1/event/61/Hands%20On%20Mahout%20-%20Mammoth%20Scale%20Machine%20Learning%20Presentation.ppt
> > >)
> > >
> > > Error: java.lang.ClassNotFoundException:
> > > org.apache.lucene.analysis.Analyzer
> > > Error: java.lang.ClassNotFoundException: org.apache.mahout.math.Vector
> > >
> > > It would be great, if someone can guide me through the specific steps
> and
> > > help me get started.
> > >
> > > Forgive me for my basic questions, i am new to mahout.
> > >
> > > Thanks & Regards,
> > >
> > > Dipesh
> > >
> > > On Sat, Nov 19, 2011 at 2:48 PM, Sean Owen <srowen@gmail.com> wrote:
> > >
> > >> You are not using the .job file, which has all the dependencies that
> you
> > >> need to send to Hadoop. I think you need to build the project.
> > >>
> > >> On Sat, Nov 19, 2011 at 3:54 AM, DIPESH KUMAR SINGH
> > >> <dipesh.tech@gmail.com>wrote:
> > >>
> > >> > Hi,
> > >> >
> > >> > I was trying to execute sample kmeans in mahout on reuters dataset
> to
> > get
> > >> > myself started with mahout. After creating the sequence files, i got
> > the
> > >> > following error.
> > >> >
> > >> > I am able to execute other map-reduce programs like wordcount on my
> > >> hadoop
> > >> > cluster.
> > >> >
> > >> > I am unable to figure how to include these missing classes which are
> > >> > indicated in exception. Please help.
> > >>  >
> > >>
> > >
> > >
> > >> >
> > >>
> > >
> > >
> > >
> > > --
> > > Dipesh Kr. Singh
> > >
> >
>
>
>
> --
> Dipesh Kr. Singh
>



-- 
Lance Norskog
goksron@gmail.com

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message