Re: Keras (abstraction layer for Theano and TensorFlow) seeks for an adopter

[Ghislain Vaillant <ghisvail@gmail.com>]

On 29/10/17 15:46, Daniel Stender wrote:
> On 10/29/2017 03:59 PM, Daniel Stender wrote:
> > Hi,
> >
> > is there somebody interested in taking up the maintenance of Keras [1] within Debian Science?
> >
> > An initial package is in experimental and appears to have some kind of an user base
> > (popcon counter is 19).
> >
> > Please adopt [2] if you like, I'll also upload the package who closes this bug report.
> >
> > Thanks,
> > Daniel Stender
> >
> > [1] https://packages.qa.debian.org/k/keras.html
> >
> > [2] https://bugs.debian.org/852133 (RFA: keras -- deep learning framework running on Theano or TensorFlow)
>
> BTW, the same is true for Lasagne [1] being up for adoption [2], which would be great
> in Debian Science, too.
>
> Best,
> DS
>
> [1] https://packages.qa.debian.org/l/lasagne.html
>
> [2] https://bugs.debian.org/852134 (RFA: lasagne -- deep learning Python library build on the top of Theano)

Considering the recent announcement that Theano's development will cease 
in the next year, I am wondering whether it is really worth spending 
efforts to maintain the Theano stack (libgpuarray, theano and lasagne)?

Would that time be better spent on the Tensorflow / Keras combo, which 
is arguably significantly more popular in research? I am afraid we are 
already spreading our packaging efforts thin on the machine learning 
ecosystem.

Ghis