Evaluator for binary classification, which expects input columns rawPrediction, label and an optional weight column.
Evaluator for clustering results.
Metrics for clustering, which expects two input columns: prediction and label.
The algorithm which is implemented in this object, instead, is an efficient and parallel implementation of the Silhouette using the cosine distance measure.
Abstract class for evaluators that compute metrics from predictions.
Evaluator for multiclass classification, which expects input columns: prediction, label, weight (optional) and probability (only for logLoss).
:: Experimental :: Evaluator for multi-label classification, which expects two input columns: prediction and label.
:: Experimental :: Evaluator for ranking, which expects two input columns: prediction and label.
Evaluator for regression, which expects input columns prediction, label and an optional weight column.
SquaredEuclideanSilhouette computes the average of the Silhouette over all the data of the dataset, which is a measure of how appropriately the data have been clustered.