[NLPL Task Force (A)] [rt.uio.no #2747687] newer versions of TensorFlow and DyNet

[Thomas Röblitz via RT]

Hei Stephan,

  we were discussing the options internally, and are wondering if you have any code example which uses both DyNet and TensorFlow at the same time? If possible we might use that for testing.

Regardless of such tests, have you decided which option mentioned below would you like to be implemented?

Kind regards

Thomas

On 2018-01-29 14:08:49, thomarob wrote:
> Hei Stephan,
> 
> have you decided which solution you want to try? I see three options
> 
> 1) DyNet and TensorFlow in one Singularity image (with an upgrade of
> TensorFlow to its last version).
> 2) DyNet and TensorFlow in separate images (with an upgrade of
> TensorFlow to its last version).
> 3) DyNet in a Singularity container, and TensorFlow outside
> Singularity (we have to figure out if that is possible, which might
> require significant time)
> 
> Kind regards
> 
> Thomas
> 
> On 2017-11-21 23:00:25, oe wrote:
> > > TensorFlow 1.3, but as explained to Aaron, it is terrible to
> > > install :)
> >
> > could you forward that earlier thread to ‘infrastructure at nlpl.eu’?
> > this is a task force comprising bjørn, joerg, martin, and myself.  we
> > would like to keep ourselves up-to-date on all technical
> > communication
> > with the infrastructure providers regarding NLPL.
> >
> > i will let aaron answer regarding the benefits of moving to the
> > current version of TensorFlow.
> >
> > with thanks in advance, oe
> >