applied ml
Heart Attack Risk Ensemble
Recall-weighted ensemble for heart attack risk, tuned so false negatives cost more than false positives.
Hugging Face Space is currently down; source on GitHub.
about
A screening model where the asymmetry matters: missing a true positive is far worse than flagging a false one, so the ensemble is tuned for recall rather than headline accuracy. Deployed as a Hugging Face Space.
stack
- scikit-learn
- Ensemble methods
- Hugging Face