Interface HasThresholds

All Superinterfaces:
Identifiable, Params, Serializable
All Known Subinterfaces:
DecisionTreeClassifierParams, FMClassifierParams, GBTClassifierParams, LogisticRegressionParams, MultilayerPerceptronParams, ProbabilisticClassifierParams, RandomForestClassifierParams, TreeEnsembleClassifierParams
All Known Implementing Classes:
Binarizer, DecisionTreeClassificationModel, DecisionTreeClassifier, FMClassificationModel, FMClassifier, GBTClassificationModel, GBTClassifier, LogisticRegression, LogisticRegressionModel, MultilayerPerceptronClassificationModel, MultilayerPerceptronClassifier, NaiveBayes, NaiveBayesModel, ProbabilisticClassificationModel, ProbabilisticClassifier, RandomForestClassificationModel, RandomForestClassifier

public interface HasThresholds extends Params
Trait for shared param thresholds. This trait may be changed or removed between minor versions.
  • Method Details

    • getThresholds

      double[] getThresholds()
    • thresholds

      DoubleArrayParam thresholds()
      Param for Thresholds in multi-class classification to adjust the probability of predicting each class. Array must have length equal to the number of classes, with values > 0 excepting that at most one value may be 0. The class with largest value p/t is predicted, where p is the original probability of that class and t is the class's threshold.
      Returns:
      (undocumented)