feat(lib): implement double-chunked ray-triangle tests for better perf.
#446Comparing
double-chunk
(b11851a
) withmain
(e36e821
)-15%
- Improvement
- 2
- Regression
- 7
- Untouched
- 6
- Archived
- 7
Benchmarks
22 total
test_transmitter_visibility[medium]
differt/tests/benchmarks/test_rt.py-52%
7 s14.6 s
test_compute_paths[medium-disconnect-hybrid]
differt/tests/benchmarks/test_rt.py-45%
30.9 s56.3 s
test_first_triangles_hit_by_rays[medium]
differt/tests/benchmarks/test_rt.py-23%
11.2 s14.5 s
test_compute_paths[small-disconnect-hybrid]
differt/tests/benchmarks/test_rt.py-22%
137.7 ms176.4 ms
test_transmitter_visibility[small]
differt/tests/benchmarks/test_rt.py-21%
32.1 ms40.9 ms
test_rays_intersect_any_triangle[medium]
differt/tests/benchmarks/test_rt.py-17%
8.5 s10.3 s
test_compute_paths[medium-no_disconnect-exhaustive]
differt/tests/benchmarks/test_rt.py-12%
20.1 s22.9 s
test_rays_intersect_any_triangle[small]
differt/tests/benchmarks/test_rt.py+84%
29 ms15.8 ms
test_first_triangles_hit_by_rays[small]
differt/tests/benchmarks/test_rt.py+40%
43 ms30.7 ms
test_fermat
differt/tests/benchmarks/test_rt.py0%
126.5 ms126.1 ms
test_compute_paths[small-disconnect-exhaustive]
differt/tests/benchmarks/test_rt.py0%
18.8 ms18.7 ms
test_compute_paths[small-no_disconnect-exhaustive]
differt/tests/benchmarks/test_rt.py0%
12 ms12 ms
test_train_step
differt/tests/benchmarks/test_training.py-2%
7.8 ms7.9 ms
test_compute_paths[medium-disconnect-exhaustive]
differt/tests/benchmarks/test_rt.py-5%
12.5 s13.2 s
test_image_method
differt/tests/benchmarks/test_rt.py-8%
157.7 µs170.5 µs
The benchmarks below were run, but are now archived.
If they were deleted in another branch, consider rebasing to remove them from the report.
If they were added back in this branch, you can restore them to include them in future performance reports.
Learn more about archiving benchmarks
If they were deleted in another branch, consider rebasing to remove them from the report.
If they were added back in this branch, you can restore them to include them in future performance reports.
Learn more about archiving benchmarks
test_compute_paths[exhaustive]
differt/tests/benchmarks/test_rt.pyArchived
0%
44.7 ms*44.7 ms*
test_compute_paths[hybrid]
differt/tests/benchmarks/test_rt.pyArchived
0%
14.5 s*14.5 s*
test_compute_paths_disconnect_inactive_triangles_benchmark[False]
differt/tests/benchmarks/test_rt.pyArchived
0%
44.9 ms*44.9 ms*
test_compute_paths_disconnect_inactive_triangles_benchmark[True]
differt/tests/benchmarks/test_rt.pyArchived
0%
30.7 ms*30.7 ms*
test_transmitter_visibility
differt/tests/benchmarks/test_rt.pyArchived
0%
2.4 s*2.4 s*
test_rays_intersect_any_triangle
differt/tests/benchmarks/test_rt.pyArchived
0%
3 s*3 s*
test_first_triangles_hit_by_rays
differt/tests/benchmarks/test_rt.pyArchived
0%
4.5 s*4.5 s*
Commits
Click on a commit to change the comparison rangeBase
main
e36e821
-12.41%
chore(docs): update changelog
7b7e197
6 days ago
by jeertmans×2.3
fix(docs): typo
0b97994
6 days ago
by jeertmansrefactor(lib): improve functions (wip)
0aad6e2
5 days ago
by jeertmans+0.3%
fix(tests): atol
4207324
5 days ago
by jeertmans×2.5
perf: trying to improve performance
d0c96bb
5 days ago
by jeertmans-81.7%
perf(lib): better batch size default
40b3121
5 days ago
by jeertmans+14.39%
cleanup
a02abdc
5 days ago
by jeertmans-23.25%
Merge remote-tracking branch 'origin/main' into double-chunk
db0d771
5 days ago
by jeertmans-68.81%
Merge branch 'main' into double-chunk
b11851a
13 hours ago
by jeertmans