Functional Efficient Python Computations

This project provides utilities for making computations (data transformations) using functional interfaces. Efficiency is both in resource consumption---no additional memory allocations, reliance on vanilla iterators, etc---and cognitive load -- we don't introduce any new syntax or metaprogramming, we stick to core python. We have a full typing annotations coverage, though at times Python limitation itself manifest itself in not being unable to guarantee correctness.

Subpackages

• ds contains useful dataclasses or contracts such as Monoid or Either,
• it provides iterator/iterable oriented functionality such as flatmaps, monoid sums or windowing,
• pa boasts with parallel processing interfaces, alternative to that of standard python's multiprocessing or concurrent.

Individual functions are documented in the docstring of the module, with illustrative examples located in tests.

All public members of the subpackages are accessible from the main package -- users are encouraged to import them from there to protect from future refactoring moves within subpackages.