Skip to content

/ nx

Multi-dimensional arrays (tensors) and numerical definitions for Elixir

1.5k stars 79 forks
Star
Notifications
main
Switch branches/tags
8 branches 0 tags
Code

Latest commit

@josevalim
josevalim Basic streaming support (#431)
28c2bc2 Aug 19, 2021
Basic streaming support (#431) 
This implements streaming for Nx default backend
and add the infrastructure for EXLA streaming (infeed,
outfeed, tokens) without implementing the defn
backend functionality.
28c2bc2

Git stats

Files

Permalink
Failed to load latest commit information.
Type
Name
Latest commit message
Commit time
.github/workflows
Run tests then check formatting
Jun 6, 2021
exla
Basic streaming support (#431)
Aug 19, 2021
nx
Basic streaming support (#431)
Aug 19, 2021
torchx
Require Erlang/OTP 24 and Elixir v1.12 (#416)
May 19, 2021
Makefile
Add Torchx README and update Makefile to run on macOS builtin make
Mar 5, 2021
README.md
Update root README
Mar 5, 2021

README.md

This repository currently holds the following projects:

  • Nx - Multi-dimensional arrays (tensors) and numerical definitions for Elixir

  • EXLA - Elixir client for Google's XLA with Nx compiler integration and JIT/AOT compilation to the CPU/GPU

  • Torchx - Elixir client to LibTorch with Nx backend integration

Each has their own README, which you can access above to learn more. They will be extracted to their own repository in the future. Examples and benchmarks are available in the EXLA project.

About

Multi-dimensional arrays (tensors) and numerical definitions for Elixir

Topics

elixir gpu jit pytorch tensor numerical xla

Resources

Readme

Contributors 32

+ 21 contributors

Languages