Package org.apache.spark.ml.param.shared
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
Trait for shared param thresholds. This trait may be changed or
 removed between minor versions.
- 
Method SummaryModifier and TypeMethodDescriptiondouble[]Param for Thresholds in multi-class classification to adjust the probability of predicting each class.Methods inherited from interface org.apache.spark.ml.util.IdentifiabletoString, uidMethods inherited from interface org.apache.spark.ml.param.Paramsclear, copy, copyValues, defaultCopy, defaultParamMap, estimateMatadataSize, explainParam, explainParams, extractParamMap, extractParamMap, get, getDefault, getOrDefault, getParam, hasDefault, hasParam, isDefined, isSet, onParamChange, paramMap, params, set, set, set, setDefault, setDefault, shouldOwn
- 
Method Details- 
getThresholdsdouble[] getThresholds()
- 
thresholdsDoubleArrayParam 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)
 
 
-