how about TF for platforms that support it and C++ for those that don't,
eventually
migrating away from C++ as support grows? a little messy for the moment
but let's
get some mileage on the TF solution ASAP
On Thu, Sep 5, 2019 at 9:21 AM Dirmeier Simon wrote:
> Hi Herve,
>
> > All this means t
I'm with Simon on the value of using TensorFlow here, and it's too bad the
build system is unable to support more flexible use of python (there are
additional challenges, for instance inter-operability between packages each
depending on & starting their own python interpreter...)
It seems like
Hi Herve,
> All this means that if you replace some old C++ code with TensorFlow
> then we will need to mark your package as unavailable on Windows and
> Mac, at least for now. Sorry. I wonder if there was something wrong
> with this old C++ code. I would recommend that you stick to it if you c
this needs to be fixed on our end, so please just be patient (I would expect a
fix later today...) Martin
On 9/5/19, 6:48 AM, "Christopher John" wrote:
Still stuck here trying to update. I think I may have been a bit special
with my version numbering.
Well let me know if the
It might be true that the code is much easier on your end, but it comes
with the complexity of using TensorFlow, which is a huge dependency. And
that library can be easy or impossible to install depending on your box. I
am not saying people shouldn't use TensorFlow, but I am saying it brings a
lot
Still stuck here trying to update. I think I may have been a bit special
with my version numbering.
Well let me know if there is a solution or do I just wait until the next
release..(?)
remote: Error: Illegal version bump from '1.7.999' to '1.7.1000'. Check
remote: http://bioconductor.org/develop
Hi Herve,
> All this means that if you replace some old C++ code with TensorFlow
> then we will need to mark your package as unavailable on Windows and
> Mac, at least for now. Sorry. I wonder if there was something wrong
> with this old C++ code. I would recommend that you stick to it if you ca
This seems like a bug in the version check code.
The next number after '999' is '1000', but '9991' is larger than '999', too, so
I think both 1000 and 9991 are valid version bumps; '10' is less than '999' so
is not a valid version bump.
Note that .99... has no special significance in the third
