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

Posted to user@spark.apache.org by Roberto Pagliari <ro...@asos.com> on 2016/02/26 02:23:22 UTC

ALS trainImplicit performance

Does anyone know about the maximum number of ratings ALS was tested successfully?

For example, is 1 billion ratings (nonzero entries) too much for it to work properly?


Thank you,

Re: ALS trainImplicit performance

Posted by Sabarish Sasidharan <sa...@manthan.com>.

I have tested upto 3 billion. ALS scales, you just need to scale your
cluster accordingly. More than building the model, it's getting the final
recommendations that won't scale as nicely, especially when number of
products is huge. This is the case when you are generating recommendations
in a batch mode and shouldn't affect if the model is used to recommend in a
real time/online mode.

Regards
Sab
On 26-Feb-2016 6:53 am, "Roberto Pagliari" <ro...@asos.com>
wrote:

> Does anyone know about the maximum number of ratings ALS was tested
> successfully?
>
> For example, is 1 billion ratings (nonzero entries) too much for it to
> work properly?
>
>
> Thank you,
>