Homogeneity#
A ground-truth clustering metric that measures the ratio of samples in a cluster that are also members of the same class. A cluster is said to be homogeneous when the entire cluster is comprised of a single class of samples.
Note
Since this metric monotonically improves as the number of target clusters increases, it should not be used as a metric to guide hyper-parameter tuning.
Estimator Compatibility: Clusterer
Output Range: 0 to 1
Parameters#
This metric does not have any parameters.
Example#
use Rubix\ML\CrossValidation\Metrics\Homogeneity;
$metric = new Homogeneity();
Last update:
2021-01-26