A ground-truth clustering metric that measures the ratio of samples in a class that are also members of the same cluster. A cluster is said to be complete when all the samples in a class are contained in a cluster.
!!! note Since this metric monotonically improves as the number of target clusters decreases, it should not be used as a metric to guide hyper-parameter tuning.
Estimator Compatibility: Clusterer
Score Range: 0 to 1
This metric does not have any parameters.
use Rubix\ML\CrossValidation\Metrics\Completeness;
$metric = new Completeness();