OpenQuant-rs

OpenQuant-rs is a Rust-first library for modern machine learning methods in finance, inspired by and building on the ideas from Advances in Financial Machine Learning by Marcos Lopez de Prado.

What this repo is

• A Rust implementation of core AFML techniques with a strict, testable baseline.
• A place to consolidate fixtures and baseline behavior before optimization.
• A stepping stone toward benchmarks and accelerated algorithms once parity is locked.

Migration status

• Source of truth: openquant-rs/tests/crosswalk.md
• Project roadmap: ROADMAP.md

Getting started

# Run all Rust tests
cargo test -p openquant

# Sync ROADMAP.md from the crosswalk
python scripts/sync_roadmap.py

Structure

• openquant-rs/crates/openquant/src/: core library modules
• openquant-rs/crates/openquant/tests/: Rust test suite
• openquant-rs/tests/fixtures/: shared fixtures used for parity
• openquant-rs/tests/crosswalk.md: mapping of Python tests to Rust tests

Principles

• Parity first: use crosswalk + shared fixtures to match Python behavior.
• Explicit tolerances: document any intentional deviations.
• Performance after parity: optimize only after tests are green and aligned.

Contributing

If you are porting a module:

1. Add or update fixtures under openquant-rs/tests/fixtures/.
2. Implement Rust tests in openquant-rs/crates/openquant/tests/.
3. Update openquant-rs/tests/crosswalk.md with status and fixtures.
4. Run python scripts/sync_roadmap.py to keep the roadmap current.

License

TBD