Open goerch opened 2 years ago
My current benchmark numbers look fine now:
(n, s) = (400, 20) jfa_voronoi! 42.500 ms (0 allocations: 0 bytes) dac_voronoi! 3.335 ms (0 allocations: 0 bytes) redac_voronoi! exact_aux = DiscreteVoronoi.exact_aux 1.216 ms (0 allocations: 0 bytes) centre_anchor_aux = DiscreteVoronoi.centre_anchor_aux 1.168 ms (0 allocations: 0 bytes) redac_voronoi_es! exact_aux_es = DiscreteVoronoi.exact_aux_es 1.142 ms (0 allocations: 0 bytes) centre_anchor_aux_es = DiscreteVoronoi.centre_anchor_aux_es 1.070 ms (0 allocations: 0 bytes) (n, s) = (400, 400) jfa_voronoi! 50.611 ms (0 allocations: 0 bytes) dac_voronoi! 260.228 ms (0 allocations: 0 bytes) redac_voronoi! exact_aux = DiscreteVoronoi.exact_aux 8.290 ms (0 allocations: 0 bytes) centre_anchor_aux = DiscreteVoronoi.centre_anchor_aux 6.158 ms (0 allocations: 0 bytes) redac_voronoi_es! exact_aux_es = DiscreteVoronoi.exact_aux_es 7.802 ms (0 allocations: 0 bytes) centre_anchor_aux_es = DiscreteVoronoi.centre_anchor_aux_es 5.792 ms (0 allocations: 0 bytes) (n, s) = (400, 8000) jfa_voronoi! 66.824 ms (0 allocations: 0 bytes) redac_voronoi! exact_aux = DiscreteVoronoi.exact_aux 204.743 ms (0 allocations: 0 bytes) centre_anchor_aux = DiscreteVoronoi.centre_anchor_aux 40.037 ms (0 allocations: 0 bytes) redac_voronoi_es! exact_aux_es = DiscreteVoronoi.exact_aux_es 199.415 ms (0 allocations: 0 bytes) centre_anchor_aux_es = DiscreteVoronoi.centre_anchor_aux_es 39.513 ms (0 allocations: 0 bytes) (n, s) = (400, 160000) jfa_voronoi! 51.920 ms (0 allocations: 0 bytes) redac_voronoi! exact_aux = DiscreteVoronoi.exact_aux 435.400 μs (0 allocations: 0 bytes) centre_anchor_aux = DiscreteVoronoi.centre_anchor_aux 432.400 μs (0 allocations: 0 bytes) redac_voronoi_es! exact_aux_es = DiscreteVoronoi.exact_aux_es 312.100 μs (0 allocations: 0 bytes) centre_anchor_aux_es = DiscreteVoronoi.centre_anchor_aux_es 313.900 μs (0 allocations: 0 bytes)
We could include preset_voronoi! into dac_voronoi! and redac_voronoi!. What do you prefer?
preset_voronoi!
dac_voronoi!
redac_voronoi!
My current benchmark numbers look fine now:
We could include
preset_voronoi!
intodac_voronoi!
andredac_voronoi!
. What do you prefer?