News for hackers. Stuff that matters.
Re:José Valim Reveals “Project Nx” (Numerical Elixir) [audio] by BadInformatics on Tuesday February 09, 2021 @12:08PM (#26079909)
Pretty disappointing to see it using XLA. Anything Tensorflow is notoriously difficult to build on a local machine and generally considered a dependency nightmare. I would've thought libtorch bindings would be much easier.
Re:José Valim Reveals “Project Nx” (Numerical Elixir) [audio] by josevalim on Tuesday February 09, 2021 @12:17PM (#26080043)
Our compiler backends are pluggable. We went with XLA because that seemed the most accessible to us 3 months ago but you should be able to bring in libtorch or any other tensor compiler.

That's actually one of the things I am really looking forward to: see what other compilers people will integrate with. I am aware some of the neural network libs have pluggable compilers but it will be interesting to see it done at the tensor level. :)

Re:José Valim Reveals “Project Nx” (Numerical Elixir) [audio] by BadInformatics on Tuesday February 09, 2021 @12:27PM (#26080182)
That's great to hear. I'd love to see the Bazel monstrosity that is Tensorflow XLA get easier to build because of Nx. Not holding my breath though!
Re:José Valim Reveals “Project Nx” (Numerical Elixir) [audio] by josevalim on Tuesday February 09, 2021 @02:56PM (#26081763)
Yeah, I would love if we could depend only on XLA. I think other communities like Julia could benefit from it too.