-
nonconform
Conformal anomaly detection and change-point detection. The library turns anomaly scores into calibrated p-values, integrates with common Python ML workflows, and supports FDR-controlled decisions through split, cross, and weighted conformal strategies.
-
online-fdr
Sequential multiple-testing procedures for online false discovery rate and family-wise error rate control. The package provides one-by-one testing APIs for procedures such as LORD, SAFFRON, ADDIS, alpha-spending, and batch baselines.
-
oddball
A lightweight collection of anomaly detection datasets for reproducible experiments.
-
aberrant
Online anomaly detection algorithms for streaming settings.
Page
Projects