We could have a module with a library of PTransforms (similar to the join
library in extensions) -- so it wouldn't be part of the core / required SDK.
On Fri, May 20, 2016 at 3:35 PM, Henry Saputra
wrote:
> I am a bit concern about adding ML model APIs to Beam because the fluctuate
> nature of
I am a bit concern about adding ML model APIs to Beam because the fluctuate
nature of ML landscape and also in reality, most data scientists tend to
use Python and R most the work with existing model definition.
Even though you could say something like Spark ML is popular, it is merely
because it
FYI: User documentation draft (the Beam Programming Guide) is well
underway. I'm regularly pushing stuff out to a fork of the Beam website
repo if anyone wants a sneak peek.
On May 20, 2016 9:37 AM, "Davor Bonaci" wrote:
> We are missing a basic getting started guide along with the rest of user
>
We are missing a basic getting started guide along with the rest of user
documentation. I think we should work on this first.
FAQ is a great idea for things that aren't or cannot be covered by those
documents -- but, we cannot really start that before we have at least a
draft version of the previo
Hi,
good idea for the FAQ. Not sure for the wiki: it would prefer kind of
governance and review using the website.
Regards
JB
On 05/20/2016 09:24 AM, Ismaël Mejía wrote:
Hello,
I have stumbled with some issues while trying to execute pipelines with all
the
different runners and I was wonder
Hello,
I have stumbled with some issues while trying to execute pipelines with all
the
different runners and I was wondering if we need to create a Frequently
Asked
Questions (FAQ) section on the website. Maybe it would be better to create
such
thing as a wiki so we can contribute faster.
What do
