Interface ProbabilisticClassificationMetric

All Superinterfaces:
Serializable
All Known Implementing Classes:
AUC, LogLoss
public interface ProbabilisticClassificationMetric extends Serializable
An abstract interface to measure the probabilistic classification performance.

  • Method Summary Link icon

    Modifier and Type
    Method
    Description
    double
    score(int[] truth, double[] probability)
    Returns a score to measure the quality of classification.

  • Method Details Link icon

    • score Link icon

      double score(int[] truth, double[] probability)
      Returns a score to measure the quality of classification.
      Parameters:
      truth - the true class labels.
      probability - The posterior probability of positive class.
      Returns:
      the metric.