2 results found Sort:

simd-everywhere
simde simd-everywhere
232
2.2k
mit
51
Implementations of SIMD instruction sets for systems which don't natively support them.
arm avx fma mmx sse avx2 gfni neon simd sse2 sse3 arm64 sse41 sse42 ssse3 avx512 altivec powerpc vectorization simd-intrinsics
Created 2017-03-28
3,019 commits to master branch, last one 9 days ago
mind
wheels mind
108
888
unknown
57
Performance-optimized wheels for TensorFlow (SSE, AVX, FMA, XLA, MPI)
ai ml avx fma gpu avx2 cuda sse41 sse42 wheel tensorflow optimization machine-learning
Created 2017-05-27
51 commits to master branch, last one 6 years ago