7 points | by Equiet 17 hours ago
2 comments
re: optimization for 1024b vectors — do you pad shorter ones, or fallback to a more general kernel?
We do a projection of the original vectors so that it matches one of our optimized kernel. This generally gives us better recall vs. simple padding since all bits are utilized.
re: optimization for 1024b vectors — do you pad shorter ones, or fallback to a more general kernel?
We do a projection of the original vectors so that it matches one of our optimized kernel. This generally gives us better recall vs. simple padding since all bits are utilized.