Models API¶
Model documentation is split by event structure and method family.
-
Static/adaptive thresholds and null/random baselines.
-
Incremental, window-replaced, half-space, Mondrian, random-cut, histogram, and xStream variants.
-
KNN, LOF, observer-distance, and point-scoring cell adaptations.
-
MStream, streaming LODA, and streaming RS-Hash adaptations.
-
AnoEdge-L, ISCONNA, MIDAS, and signed graph sketches.
-
Pure-online X-Lag Amnesic DAMP.
-
Candidate-induced changes in moving univariate/bivariate statistics and squared Mahalanobis distance.
-
Experimental budgeted and graph-gated one-class heuristics.
-
NumPy online autoencoder ensemble and optional user-supplied PyTorch autoencoder.
All model classes expose learn_one and score_one, but their event
schemas, readiness conditions, state bounds, and score scales differ. Start
with the model guide when selecting a class.