Package org.apache.spark.ml.tree
Interface GBTRegressorParams
- All Superinterfaces:
- DecisionTreeParams,- GBTParams,- HasCheckpointInterval,- HasFeaturesCol,- HasLabelCol,- HasMaxIter,- HasPredictionCol,- HasSeed,- HasStepSize,- HasValidationIndicatorCol,- HasVarianceImpurity,- HasWeightCol,- Identifiable,- Params,- PredictorParams,- Serializable,- TreeEnsembleParams,- TreeEnsembleRegressorParams,- TreeRegressorParams
- All Known Implementing Classes:
- GBTRegressionModel,- GBTRegressor
public interface GBTRegressorParams
extends GBTParams, TreeEnsembleRegressorParams, TreeRegressorParams
- 
Method SummaryModifier and TypeMethodDescriptionconvertToOldLossType(String loss) (private[ml]) Convert new loss to old loss.lossType()Loss function which GBT tries to minimize.Methods inherited from interface org.apache.spark.ml.tree.DecisionTreeParamscacheNodeIds, getCacheNodeIds, getLeafCol, getMaxBins, getMaxDepth, getMaxMemoryInMB, getMinInfoGain, getMinInstancesPerNode, getMinWeightFractionPerNode, getOldStrategy, leafCol, maxBins, maxDepth, maxMemoryInMB, minInfoGain, minInstancesPerNode, minWeightFractionPerNode, setLeafColMethods inherited from interface org.apache.spark.ml.tree.GBTParamsgetOldBoostingStrategy, getValidationTol, stepSize, validationTolMethods inherited from interface org.apache.spark.ml.param.shared.HasCheckpointIntervalcheckpointInterval, getCheckpointIntervalMethods inherited from interface org.apache.spark.ml.param.shared.HasFeaturesColfeaturesCol, getFeaturesColMethods inherited from interface org.apache.spark.ml.param.shared.HasLabelColgetLabelCol, labelColMethods inherited from interface org.apache.spark.ml.param.shared.HasMaxItergetMaxIter, maxIterMethods inherited from interface org.apache.spark.ml.param.shared.HasPredictionColgetPredictionCol, predictionColMethods inherited from interface org.apache.spark.ml.param.shared.HasStepSizegetStepSizeMethods inherited from interface org.apache.spark.ml.param.shared.HasValidationIndicatorColgetValidationIndicatorCol, validationIndicatorColMethods inherited from interface org.apache.spark.ml.tree.HasVarianceImpuritygetImpurity, getOldImpurity, impurityMethods inherited from interface org.apache.spark.ml.param.shared.HasWeightColgetWeightCol, weightColMethods 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, shouldOwnMethods inherited from interface org.apache.spark.ml.tree.TreeEnsembleParamsfeatureSubsetStrategy, getFeatureSubsetStrategy, getOldStrategy, getSubsamplingRate, subsamplingRateMethods inherited from interface org.apache.spark.ml.tree.TreeEnsembleRegressorParamsvalidateAndTransformSchema
- 
Method Details- 
lossTypeLoss function which GBT tries to minimize. (case-insensitive) Supported: "squared" (L2) and "absolute" (L1) (default = squared)- Returns:
- (undocumented)
 
- 
getLossTypeString getLossType()
- 
getOldLossTypeLoss getOldLossType()(private[ml]) Convert new loss to old loss.- Specified by:
- getOldLossTypein interface- GBTParams
 
- 
convertToOldLossType
 
-