Benchmarks

Demonstration checks

Documentation · Tutorial

The bundled synthetic example demonstrates KEA, KRS and K-RBP with small inputs and fixed seeds. It is a software check, not a measure of biological prediction accuracy.

python examples/run_demo.py --out results/demo

Demo outputs are in examples/demo_output:

• KEA recovers the injected ACG repeat signature (ACG, CGA, GAC).
• KRS selects eight words for the example target at top_pct=10.
• K-RBP evaluates three fictional profiles and assigns a positive Cliff's delta to the query-favoring profile.

The regression tests cover reference counting boundaries and overlaps, KRS ranks and ties, K-RBP effect direction, K-map, K-CDS and miRNA seed orientation:

python -m pip install -e '.[dev]'
python -m pytest