ml-practice

A workspace structured project to help revise machine learning algorithms, maturin-based Rust extension modules, and experiment with uv and its monorepo.

The top-level workspace defines the development dependencies and any cross-cutting tests. Underlying this there are 3 packages:

1. py_algos: this is a pure Python package for machine learning algorithm implementations. Generally, this will include both a native version (using only standard library objects), and a numpy-based version with matrix operations.
2. rs_algos: this is a Rust extension package, with parallel machine learning algorithm implementations exposed to Python through maturin/PyO3.
3. ml_api: this is a FastAPI server, to provide a convenient interface to test each of the 3 parallel implementations for a given algorithm.