Various performance improvements#83

Merged

r-devulap merged 24 commits intonumpy:mainfrom

sterrettm2:perfwork

Oct 13, 2023

Commits on Oct 10, 2023

Commits on Oct 11, 2023

Changed how partitioning small arrays is handled
sterrettm2
committed
Add new CI to test parition_avx512 code
Raghuveer Devulapalli
committed

Commits on Oct 12, 2023

Fixed minor bug in partition code
sterrettm2
committed
Fix formatting and add a few comments
Raghuveer Devulapalli
committed
Fix more formatting
Raghuveer Devulapalli
committed
Use num_unroll = 9 for 16-bit data
Raghuveer Devulapalli
committed
Do not use a set seed in tests
Raghuveer Devulapalli
committed

Commits on Oct 13, 2023

Add comments
Raghuveer Devulapalli
committed
Bug fix in tests
Raghuveer Devulapalli
committed