approximate_predict

fast_plscan.prediction.approximate_predict(clusterer, X)

Approximate labels and membership probabilities for unseen points.

This follows HDBSCAN*-style approximate prediction: each new point is connected to the fitted structure through its nearest mutual-reachability neighbor in the training set.

Parameters:

• clusterer (PLSCAN) – A fitted PLSCAN estimator trained on feature-vector input.

• X (ndarray[tuple[int, int], dtype[single]]) – New feature vectors with shape (n_samples_new, n_features).

Return type:

tuple[ndarray[tuple[int], dtype[int_]], ndarray[tuple[int], dtype[single]]]

Returns:

• labels – Predicted cluster labels for each new point. Points that cannot be linked to a selected cluster are labeled -1.

• probabilities – Approximate membership probabilities in [0, 1] for each new point.

Raises:

• NotFittedError – If clusterer has not been fitted.

• ValueError – If clusterer was fitted with precomputed input, or if X has an invalid number of features.