4 ms·
Note however, that on Intel it's actually slower than run off the mill float32 linear algebra library like Eigen or OpenBLAS. Its main forte seems to be ARM.
by throwaway71958 9y ago
Note however, that on Intel it's actually slower than run off the mill float32 linear algebra library like Eigen or OpenBLAS. Its main forte seems to be ARM.