Hello David, Sorry for the late response. The bindings use the C library (not C++). Tensorflow has a Protocol Buffer description of the bindings. Using it one can generate bindings with relatively little effort. I don't have a GPU. So I only have tested it using CPUs.
Regards Jan On 15 December 2018 22:55:06 GMT+00:00, David Pirotte <da...@altosw.be> wrote: >Hello Jan, > >> The current AIscm source now include Tensorflow bindings [1]. >> This means that one can now develop machine learning software using >Guile >> and Tensorflow. More examples to follow ... > >This is excellent news! > >I've had no time (yet), but I'm interested to see what approach you >used to bind it >(you did bind the C++ lib right?), and how aiscm leads with errors (C++ >exceptions) > >With proper config, I guess it uses all available GPUs? > >David -- Diese Nachricht wurde von meinem Android-Gerät mit K-9 Mail gesendet.
