Pls check this link 
http://shuyo.wordpress.com/2011/02/14/mahout-development-environment-with-maven-and-eclipse-2/

Here mahout is being run from eclipse without needing Hadoop infra. 

I think Mahout can be used as a library too. Some algorithms are implemented in 
map-reduce fashion and they may need Hadoop, but rewriting them in non-hadoop 
fashion should not be a big deal.



-----Original Message-----
From: Marek Slim [mailto:[email protected]] 
Sent: Thursday, August 16, 2012 9:27 PM
To: [email protected]
Subject: Re: Apache Mahout without Hadoop

I read that Mahout has very good text classifier and licence is by Apache
(not for example GPL like Weka).
I don't need (now) features of Hadoop. I just want to use text classifiers.
I heard that setting up all Hadoop is quite painful.. that why I want to
use just Mahout.

Do you think that setting up MAHOUR_LOCAL is enough ?

2012/8/15 Arun Ahuja <[email protected]>

> Is there a particular reason you want to do this?  There are other
> tools out there that will probably be simpler to use if you *only*
> want to run locally without Hadoop.
>
> However, if you set MAHOUT_LOCAL environment variable you should be
> able to run any of the programs locally.
>
> Arun
>
> On Tue, Aug 14, 2012 at 10:08 AM, Marek Slim <[email protected]> wrote:
> > Hi,
> >
> > I would like to use text classifier tool, that are implemented in Apache
> > Mahout.
> > I don't want to run Mahout with Apache Hadoop. I would like to use
> > classification in simple way.
> >
> > Is it possible to run Apache Mahout independently without Haddop ?
> > Would I be able to use text classifiers ?
> >
> > Thanks for help
>

Reply via email to