XNNPACK

Description:library of floating-point neural network inference operators
License: MIT
Homepage:https://github.com/google/XNNPACK/
Location:Portage
Legend:
  • + - stable
  • ~ - unstable
  • M - hardmask
alpha amd64 arm hppa ia64 m68k mips ppc ppc64 s390 sh sparc x86
2022.12.22 diff - ~ - - - - - - - - - - -
2022.02.17-r1 - ~ - - - - - - - - - - -

USE flags

  • assembly - Build with assembly micro-kernels
  • jit - Build with JIT micro-kernels
  • memopt - Build with optimized memory allocation scheme
  • sparse - Build with graph rewriting for sparse inference
  • static-libs - Build static versions of dynamic libraries as well
  • test - Enable dependencies and/or preparations necessary to run tests (usually controlled by FEATURES=test but can be toggled independently)

Also available in: Atom

Thank you!