Roadmap

Version 2.0 of the package uses multithreaded ufunc loops and parallel sorts.

Future versions of the package will extend these capabilites to cover more of the NumPy functionality. Some of these proposed enhancements will require new APIs from NumPy.

Conversions

Currently NumPy does not expose a hook for dtype conversions. When available, PNumPy will parallelize those conversions.

Vectorized loops

NumPy is only now beginning to use SIMD instructions to speed up loops. We have a few further enhancements to the current NumPy implementations. Check out the code in the atop directory.

Using a better memory allocator

NumPy uses a small cache for data memory but does not have one for larger arrays. When the new API is available, we will provide a better cache.

Ledger

What PNumPy hooks can be recorded and timed. This built in profiler will help you to tweak and speed up your code.