Closed Spaceenter closed 7 years ago
@babbush
The motivation for this improvement is that, through cProfile, we found the bottleneck is computing momentum_vectors too many times.
@babbush