dataeval.core.compute_cluster_stats¶

dataeval.core.compute_cluster_stats(embeddings, clusters)¶

Compute cluster centers and distance statistics for adaptive outlier detection.

Parameters:¶

embeddings : NDArray[np.floating]¶

The embedding vectors, shape (n_samples, n_features)

clusters : NDArray[np.int64]¶

Cluster labels from HDBSCAN (-1 for HDBSCAN outliers)

Returns:¶

Pre-calculated statistics with empty arrays if no valid clusters found

Return type:¶

ClusterStats