You are viewing a plain text version of this content. The canonical link for it is here.

Posted to user@mahout.apache.org by Mohammed Al khooja <mk...@gmail.com> on 2011/11/29 03:38:50 UTC

LDA vocabulary limits

Hi,

What is the limit for the vocabulary that can be used in Mahout LDA ?  It's
crashing when I use 340,000 vocab terms.

Thanks.

-- 

M.khouja

Re: LDA vocabulary limits

Posted by Omkar Raut <om...@gmail.com>.

Hi

Sorry to post another question on top, but are you able to print say "*top
10 topics in a given document*" using a smaller set of vocab items?

Not sure if you need to specify the max vocab items as parameter, you can
try increasing that.. there could be a default max value set.

Thanks,
Omkar

On Mon, Nov 28, 2011 at 6:38 PM, Mohammed Al khooja <mk...@gmail.com>wrote:

> Hi,
>
> What is the limit for the vocabulary that can be used in Mahout LDA ?  It's
> crashing when I use 340,000 vocab terms.
>
> Thanks.
>
> --
>
> M.khouja
>

Re: LDA vocabulary limits

Posted by Jake Mannix <ja...@gmail.com>.

the following must fit in RAM in each Mapper instance: numTerms x numTopics
x 8bytes

  -jake

On Mon, Nov 28, 2011 at 6:38 PM, Mohammed Al khooja <mk...@gmail.com>wrote:

> Hi,
>
> What is the limit for the vocabulary that can be used in Mahout LDA ?  It's
> crashing when I use 340,000 vocab terms.
>
> Thanks.
>
> --
>
> M.khouja
>