Hi Marko, Nothing special needs to be done to get Mahout 0.9 running on Mahout 0.2. The steps are same as 1.2.1. You might refer to this link http://chandramanitiwary.wordpress.com/2014/03/20/installing-apache-mahout-on-ubuntu/ and there are other good links to for the same. I don't remember the dependencies, I would advise you to do sanity testing once you have mahout configured.
Regards Chandramani On Tue, Oct 28, 2014 at 9:28 PM, Marko Dinić <[email protected]> wrote: > Hi Chandramani, > > Thank you very much for your answer. > > Can you please explain how can I do the configuration? > > And also, in which amount can expect failures? I'm currently using > clustering algorithms, concretely Streaming K Means, do you perhaps have > any experience with it, should I expect failures in case of it? > > Regards, > Marko > > > On уторак, 28. октобар 2014. 16:48:03 CET, Chandramani Tiwary wrote: > >> Hi Marko, >> >> You can configure Mahout 0.9 over Hadoop 0.20.2 but the Hadoop >> dependencies >> might lead to failure quite a few time. One example, If I remember >> correctly is that Hadoop 0.20.2 doesn't have MultipleInputs which is >> required by a few algorithms of Mahout 0.9. >> >> Regards >> Chandramani >> >> On Tue, Oct 28, 2014 at 8:05 PM, Marko Dinić <[email protected]> >> wrote: >> >> Hello, >>> >>> I have Hadoop cluster on which Hadoop 0.20.2 is installed. Is there a way >>> to use Mahout 0.9 on that cluster? >>> >>> I understand that Mahout 0.9 is based on Hadoop 1.2.1, but I have this >>> constraint, so I cannot install another version of Hadoop on it. >>> >>> Thanks, >>> Marko >>> >>> >> > -- > Pozdrav, > Marko Dinić >
