-
Notifications
You must be signed in to change notification settings - Fork 0
Performance
Wolren edited this page Apr 28, 2026
·
5 revisions
Benchmarks run on i5-12400F, 32GB DDR4, default parameters, Numba installed. 290 and 5406 are the number of features in the testing dataset.
| Algorithm | Time @290 (s) | Time @5406 (s) | Best Mode |
|---|---|---|---|
| Approx Fast | 6.98 | 125.93 | 12w+chunk |
| Axis-Aligned | 11.81 | 120.24 | 1w |
| Contained Fast | 12.25 | 226.05 | 12w+chunk |
| BCRS Fast | 23.61 | 445.01 | 1w |
| Profile | Algorithm | Time @290 (s) | Time @5406 (s) | Scale |
|---|---|---|---|---|
| P1 | Approx Standard | 7.13 | 127.25 | 17.85x |
| P2 | Approx Fast | 6.98 | 125.93 | 18.04x |
| P5 | Contained Fast | 12.25 | 226.05 | 18.45x |
| P8 | BCRS Fast | 23.61 | 445.01 | 18.85x |
| P9 | Axis-Aligned | 11.81 | 120.24 | 10.18x |
- Best throughput: Approx Fast (6.98s @290)
- Best scaling: Axis-Aligned (10.18x vs ~18x for others)
- Most accurate: BCRS Fast for rotated rectangles
- BCRS/Axis-Aligned: Negative speedup with multithreading
See Complexity for formal analysis.
Navigation