LightGBMRanker implements LightGBMRanker More...

Inheritance diagram for Synapse.ML.Lightgbm.LightGBMRanker:

[legend]

Collaboration diagram for Synapse.ML.Lightgbm.LightGBMRanker:

[legend]

Public Member Functions
	LightGBMRanker ()
	Creates a LightGBMRanker without any parameters. More...

	LightGBMRanker (string uid)
	Creates a LightGBMRanker with a UID that is used to give the LightGBMRanker a unique ID. More...

LightGBMRanker	SetBaggingFraction (double value)
	Sets value for baggingFraction More...

LightGBMRanker	SetBaggingFreq (int value)
	Sets value for baggingFreq More...

LightGBMRanker	SetBaggingSeed (int value)
	Sets value for baggingSeed More...

LightGBMRanker	SetBinSampleCount (int value)
	Sets value for binSampleCount More...

LightGBMRanker	SetBoostFromAverage (bool value)
	Sets value for boostFromAverage More...

LightGBMRanker	SetBoostingType (string value)
	Sets value for boostingType More...

LightGBMRanker	SetCatSmooth (double value)
	Sets value for catSmooth More...

LightGBMRanker	SetCategoricalSlotIndexes (int[] value)
	Sets value for categoricalSlotIndexes More...

LightGBMRanker	SetCategoricalSlotNames (string[] value)
	Sets value for categoricalSlotNames More...

LightGBMRanker	SetCatl2 (double value)
	Sets value for catl2 More...

LightGBMRanker	SetChunkSize (int value)
	Sets value for chunkSize More...

LightGBMRanker	SetDataRandomSeed (int value)
	Sets value for dataRandomSeed More...

LightGBMRanker	SetDataTransferMode (string value)
	Sets value for dataTransferMode More...

LightGBMRanker	SetDefaultListenPort (int value)
	Sets value for defaultListenPort More...

LightGBMRanker	SetDeterministic (bool value)
	Sets value for deterministic More...

LightGBMRanker	SetDriverListenPort (int value)
	Sets value for driverListenPort More...

LightGBMRanker	SetDropRate (double value)
	Sets value for dropRate More...

LightGBMRanker	SetDropSeed (int value)
	Sets value for dropSeed More...

LightGBMRanker	SetEarlyStoppingRound (int value)
	Sets value for earlyStoppingRound More...

LightGBMRanker	SetEvalAt (int[] value)
	Sets value for evalAt More...

LightGBMRanker	SetExecutionMode (string value)
	Sets value for executionMode More...

LightGBMRanker	SetExtraSeed (int value)
	Sets value for extraSeed More...

LightGBMRanker	SetFeatureFraction (double value)
	Sets value for featureFraction More...

LightGBMRanker	SetFeatureFractionByNode (double value)
	Sets value for featureFractionByNode More...

LightGBMRanker	SetFeatureFractionSeed (int value)
	Sets value for featureFractionSeed More...

LightGBMRanker	SetFeaturesCol (string value)
	Sets value for featuresCol More...

LightGBMRanker	SetFeaturesShapCol (string value)
	Sets value for featuresShapCol More...

LightGBMRanker	SetFobj (object value)
	Sets value for fobj More...

LightGBMRanker	SetGroupCol (string value)
	Sets value for groupCol More...

LightGBMRanker	SetImprovementTolerance (double value)
	Sets value for improvementTolerance More...

LightGBMRanker	SetInitScoreCol (string value)
	Sets value for initScoreCol More...

LightGBMRanker	SetIsEnableSparse (bool value)
	Sets value for isEnableSparse More...

LightGBMRanker	SetIsProvideTrainingMetric (bool value)
	Sets value for isProvideTrainingMetric More...

LightGBMRanker	SetLabelCol (string value)
	Sets value for labelCol More...

LightGBMRanker	SetLabelGain (double[] value)
	Sets value for labelGain More...

LightGBMRanker	SetLambdaL1 (double value)
	Sets value for lambdaL1 More...

LightGBMRanker	SetLambdaL2 (double value)
	Sets value for lambdaL2 More...

LightGBMRanker	SetLeafPredictionCol (string value)
	Sets value for leafPredictionCol More...

LightGBMRanker	SetLearningRate (double value)
	Sets value for learningRate More...

LightGBMRanker	SetMatrixType (string value)
	Sets value for matrixType More...

LightGBMRanker	SetMaxBin (int value)
	Sets value for maxBin More...

LightGBMRanker	SetMaxBinByFeature (int[] value)
	Sets value for maxBinByFeature More...

LightGBMRanker	SetMaxCatThreshold (int value)
	Sets value for maxCatThreshold More...

LightGBMRanker	SetMaxCatToOnehot (int value)
	Sets value for maxCatToOnehot More...

LightGBMRanker	SetMaxDeltaStep (double value)
	Sets value for maxDeltaStep More...

LightGBMRanker	SetMaxDepth (int value)
	Sets value for maxDepth More...

LightGBMRanker	SetMaxDrop (int value)
	Sets value for maxDrop More...

LightGBMRanker	SetMaxNumClasses (int value)
	Sets value for maxNumClasses More...

LightGBMRanker	SetMaxPosition (int value)
	Sets value for maxPosition More...

LightGBMRanker	SetMaxStreamingOMPThreads (int value)
	Sets value for maxStreamingOMPThreads More...

LightGBMRanker	SetMetric (string value)
	Sets value for metric More...

LightGBMRanker	SetMicroBatchSize (int value)
	Sets value for microBatchSize More...

LightGBMRanker	SetMinDataInLeaf (int value)
	Sets value for minDataInLeaf More...

LightGBMRanker	SetMinDataPerBin (int value)
	Sets value for minDataPerBin More...

LightGBMRanker	SetMinDataPerGroup (int value)
	Sets value for minDataPerGroup More...

LightGBMRanker	SetMinGainToSplit (double value)
	Sets value for minGainToSplit More...

LightGBMRanker	SetMinSumHessianInLeaf (double value)
	Sets value for minSumHessianInLeaf More...

LightGBMRanker	SetModelString (string value)
	Sets value for modelString More...

LightGBMRanker	SetMonotoneConstraints (int[] value)
	Sets value for monotoneConstraints More...

LightGBMRanker	SetMonotoneConstraintsMethod (string value)
	Sets value for monotoneConstraintsMethod More...

LightGBMRanker	SetMonotonePenalty (double value)
	Sets value for monotonePenalty More...

LightGBMRanker	SetNegBaggingFraction (double value)
	Sets value for negBaggingFraction More...

LightGBMRanker	SetNumBatches (int value)
	Sets value for numBatches More...

LightGBMRanker	SetNumIterations (int value)
	Sets value for numIterations More...

LightGBMRanker	SetNumLeaves (int value)
	Sets value for numLeaves More...

LightGBMRanker	SetNumTasks (int value)
	Sets value for numTasks More...

LightGBMRanker	SetNumThreads (int value)
	Sets value for numThreads More...

LightGBMRanker	SetObjective (string value)
	Sets value for objective More...

LightGBMRanker	SetObjectiveSeed (int value)
	Sets value for objectiveSeed More...

LightGBMRanker	SetOtherRate (double value)
	Sets value for otherRate More...

LightGBMRanker	SetParallelism (string value)
	Sets value for parallelism More...

LightGBMRanker	SetPassThroughArgs (string value)
	Sets value for passThroughArgs More...

LightGBMRanker	SetPosBaggingFraction (double value)
	Sets value for posBaggingFraction More...

LightGBMRanker	SetPredictDisableShapeCheck (bool value)
	Sets value for predictDisableShapeCheck More...

LightGBMRanker	SetPredictionCol (string value)
	Sets value for predictionCol More...

LightGBMRanker	SetReferenceDataset (object value)
	Sets value for referenceDataset More...

LightGBMRanker	SetRepartitionByGroupingColumn (bool value)
	Sets value for repartitionByGroupingColumn More...

LightGBMRanker	SetSamplingMode (string value)
	Sets value for samplingMode More...

LightGBMRanker	SetSamplingSubsetSize (int value)
	Sets value for samplingSubsetSize More...

LightGBMRanker	SetSeed (int value)
	Sets value for seed More...

LightGBMRanker	SetSkipDrop (double value)
	Sets value for skipDrop More...

LightGBMRanker	SetSlotNames (string[] value)
	Sets value for slotNames More...

LightGBMRanker	SetTimeout (double value)
	Sets value for timeout More...

LightGBMRanker	SetTopK (int value)
	Sets value for topK More...

LightGBMRanker	SetTopRate (double value)
	Sets value for topRate More...

LightGBMRanker	SetUniformDrop (bool value)
	Sets value for uniformDrop More...

LightGBMRanker	SetUseBarrierExecutionMode (bool value)
	Sets value for useBarrierExecutionMode More...

LightGBMRanker	SetUseMissing (bool value)
	Sets value for useMissing More...

LightGBMRanker	SetUseSingleDatasetMode (bool value)
	Sets value for useSingleDatasetMode More...

LightGBMRanker	SetValidationIndicatorCol (string value)
	Sets value for validationIndicatorCol More...

LightGBMRanker	SetVerbosity (int value)
	Sets value for verbosity More...

LightGBMRanker	SetWeightCol (string value)
	Sets value for weightCol More...

LightGBMRanker	SetXGBoostDartMode (bool value)
	Sets value for xGBoostDartMode More...

LightGBMRanker	SetZeroAsMissing (bool value)
	Sets value for zeroAsMissing More...

double	GetBaggingFraction ()
	Gets baggingFraction value More...

int	GetBaggingFreq ()
	Gets baggingFreq value More...

int	GetBaggingSeed ()
	Gets baggingSeed value More...

int	GetBinSampleCount ()
	Gets binSampleCount value More...

bool	GetBoostFromAverage ()
	Gets boostFromAverage value More...

string	GetBoostingType ()
	Gets boostingType value More...

double	GetCatSmooth ()
	Gets catSmooth value More...

int[]	GetCategoricalSlotIndexes ()
	Gets categoricalSlotIndexes value More...

string[]	GetCategoricalSlotNames ()
	Gets categoricalSlotNames value More...

double	GetCatl2 ()
	Gets catl2 value More...

int	GetChunkSize ()
	Gets chunkSize value More...

int	GetDataRandomSeed ()
	Gets dataRandomSeed value More...

string	GetDataTransferMode ()
	Gets dataTransferMode value More...

int	GetDefaultListenPort ()
	Gets defaultListenPort value More...

bool	GetDeterministic ()
	Gets deterministic value More...

int	GetDriverListenPort ()
	Gets driverListenPort value More...

double	GetDropRate ()
	Gets dropRate value More...

int	GetDropSeed ()
	Gets dropSeed value More...

int	GetEarlyStoppingRound ()
	Gets earlyStoppingRound value More...

int[]	GetEvalAt ()
	Gets evalAt value More...

string	GetExecutionMode ()
	Gets executionMode value More...

int	GetExtraSeed ()
	Gets extraSeed value More...

double	GetFeatureFraction ()
	Gets featureFraction value More...

double	GetFeatureFractionByNode ()
	Gets featureFractionByNode value More...

int	GetFeatureFractionSeed ()
	Gets featureFractionSeed value More...

string	GetFeaturesCol ()
	Gets featuresCol value More...

string	GetFeaturesShapCol ()
	Gets featuresShapCol value More...

object	GetFobj ()
	Gets fobj value More...

string	GetGroupCol ()
	Gets groupCol value More...

double	GetImprovementTolerance ()
	Gets improvementTolerance value More...

string	GetInitScoreCol ()
	Gets initScoreCol value More...

bool	GetIsEnableSparse ()
	Gets isEnableSparse value More...

bool	GetIsProvideTrainingMetric ()
	Gets isProvideTrainingMetric value More...

string	GetLabelCol ()
	Gets labelCol value More...

double[]	GetLabelGain ()
	Gets labelGain value More...

double	GetLambdaL1 ()
	Gets lambdaL1 value More...

double	GetLambdaL2 ()
	Gets lambdaL2 value More...

string	GetLeafPredictionCol ()
	Gets leafPredictionCol value More...

double	GetLearningRate ()
	Gets learningRate value More...

string	GetMatrixType ()
	Gets matrixType value More...

int	GetMaxBin ()
	Gets maxBin value More...

int[]	GetMaxBinByFeature ()
	Gets maxBinByFeature value More...

int	GetMaxCatThreshold ()
	Gets maxCatThreshold value More...

int	GetMaxCatToOnehot ()
	Gets maxCatToOnehot value More...

double	GetMaxDeltaStep ()
	Gets maxDeltaStep value More...

int	GetMaxDepth ()
	Gets maxDepth value More...

int	GetMaxDrop ()
	Gets maxDrop value More...

int	GetMaxNumClasses ()
	Gets maxNumClasses value More...

int	GetMaxPosition ()
	Gets maxPosition value More...

int	GetMaxStreamingOMPThreads ()
	Gets maxStreamingOMPThreads value More...

string	GetMetric ()
	Gets metric value More...

int	GetMicroBatchSize ()
	Gets microBatchSize value More...

int	GetMinDataInLeaf ()
	Gets minDataInLeaf value More...

int	GetMinDataPerBin ()
	Gets minDataPerBin value More...

int	GetMinDataPerGroup ()
	Gets minDataPerGroup value More...

double	GetMinGainToSplit ()
	Gets minGainToSplit value More...

double	GetMinSumHessianInLeaf ()
	Gets minSumHessianInLeaf value More...

string	GetModelString ()
	Gets modelString value More...

int[]	GetMonotoneConstraints ()
	Gets monotoneConstraints value More...

string	GetMonotoneConstraintsMethod ()
	Gets monotoneConstraintsMethod value More...

double	GetMonotonePenalty ()
	Gets monotonePenalty value More...

double	GetNegBaggingFraction ()
	Gets negBaggingFraction value More...

int	GetNumBatches ()
	Gets numBatches value More...

int	GetNumIterations ()
	Gets numIterations value More...

int	GetNumLeaves ()
	Gets numLeaves value More...

int	GetNumTasks ()
	Gets numTasks value More...

int	GetNumThreads ()
	Gets numThreads value More...

string	GetObjective ()
	Gets objective value More...

int	GetObjectiveSeed ()
	Gets objectiveSeed value More...

double	GetOtherRate ()
	Gets otherRate value More...

string	GetParallelism ()
	Gets parallelism value More...

string	GetPassThroughArgs ()
	Gets passThroughArgs value More...

double	GetPosBaggingFraction ()
	Gets posBaggingFraction value More...

bool	GetPredictDisableShapeCheck ()
	Gets predictDisableShapeCheck value More...

string	GetPredictionCol ()
	Gets predictionCol value More...

object	GetReferenceDataset ()
	Gets referenceDataset value More...

bool	GetRepartitionByGroupingColumn ()
	Gets repartitionByGroupingColumn value More...

string	GetSamplingMode ()
	Gets samplingMode value More...

int	GetSamplingSubsetSize ()
	Gets samplingSubsetSize value More...

int	GetSeed ()
	Gets seed value More...

double	GetSkipDrop ()
	Gets skipDrop value More...

string[]	GetSlotNames ()
	Gets slotNames value More...

double	GetTimeout ()
	Gets timeout value More...

int	GetTopK ()
	Gets topK value More...

double	GetTopRate ()
	Gets topRate value More...

bool	GetUniformDrop ()
	Gets uniformDrop value More...

bool	GetUseBarrierExecutionMode ()
	Gets useBarrierExecutionMode value More...

bool	GetUseMissing ()
	Gets useMissing value More...

bool	GetUseSingleDatasetMode ()
	Gets useSingleDatasetMode value More...

string	GetValidationIndicatorCol ()
	Gets validationIndicatorCol value More...

int	GetVerbosity ()
	Gets verbosity value More...

string	GetWeightCol ()
	Gets weightCol value More...

bool	GetXGBoostDartMode ()
	Gets xGBoostDartMode value More...

bool	GetZeroAsMissing ()
	Gets zeroAsMissing value More...

override LightGBMRankerModel	Fit (DataFrame dataset)
	Fits a model to the input data. More...

void	Save (string path)
	Saves the object so that it can be loaded later using Load. Note that these objects can be shared with Scala by Loading or Saving in Scala. More...

JavaMLWriter	Write ()
	Returns a JavaMLWriter instance for this ML instance.

JavaMLReader< LightGBMRanker >	Read ()
	Get the corresponding JavaMLReader instance. More...

Static Public Member Functions
static LightGBMRanker	Load (string path)
	Loads the LightGBMRanker that was previously saved using Save(string). More...

Detailed Description

LightGBMRanker implements LightGBMRanker

Constructor & Destructor Documentation

◆ LightGBMRanker() [1/2]

Synapse.ML.Lightgbm.LightGBMRanker.LightGBMRanker ( )

inline

Creates a LightGBMRanker without any parameters.

◆ LightGBMRanker() [2/2]

Synapse.ML.Lightgbm.LightGBMRanker.LightGBMRanker ( string uid )

inline

Creates a LightGBMRanker with a UID that is used to give the LightGBMRanker a unique ID.

Parameters

uid	An immutable unique ID for the object and its derivatives.

Member Function Documentation

◆ Fit()

override LightGBMRankerModel Synapse.ML.Lightgbm.LightGBMRanker.Fit ( DataFrame dataset )

Fits a model to the input data.

Parameters

dataset The DataFrame to fit the model to.

Returns: LightGBMRankerModel

◆ GetBaggingFraction()

double Synapse.ML.Lightgbm.LightGBMRanker.GetBaggingFraction ( )

Gets baggingFraction value

Returns: baggingFraction: Bagging fraction

◆ GetBaggingFreq()

int Synapse.ML.Lightgbm.LightGBMRanker.GetBaggingFreq ( )

Gets baggingFreq value

Returns: baggingFreq: Bagging frequency

◆ GetBaggingSeed()

int Synapse.ML.Lightgbm.LightGBMRanker.GetBaggingSeed ( )

Gets baggingSeed value

Returns: baggingSeed: Bagging seed

◆ GetBinSampleCount()

int Synapse.ML.Lightgbm.LightGBMRanker.GetBinSampleCount ( )

Gets binSampleCount value

Returns: binSampleCount: Number of samples considered at computing histogram bins

◆ GetBoostFromAverage()

bool Synapse.ML.Lightgbm.LightGBMRanker.GetBoostFromAverage ( )

Gets boostFromAverage value

Returns: boostFromAverage: Adjusts initial score to the mean of labels for faster convergence

◆ GetBoostingType()

string Synapse.ML.Lightgbm.LightGBMRanker.GetBoostingType ( )

Gets boostingType value

Returns: boostingType: Default gbdt = traditional Gradient Boosting Decision Tree. Options are: gbdt, gbrt, rf (Random Forest), random_forest, dart (Dropouts meet Multiple Additive Regression Trees), goss (Gradient-based One-Side Sampling).

◆ GetCategoricalSlotIndexes()

int [] Synapse.ML.Lightgbm.LightGBMRanker.GetCategoricalSlotIndexes ( )

Gets categoricalSlotIndexes value

Returns: categoricalSlotIndexes: List of categorical column indexes, the slot index in the features column

◆ GetCategoricalSlotNames()

string [] Synapse.ML.Lightgbm.LightGBMRanker.GetCategoricalSlotNames ( )

Gets categoricalSlotNames value

Returns: categoricalSlotNames: List of categorical column slot names, the slot name in the features column

◆ GetCatl2()

double Synapse.ML.Lightgbm.LightGBMRanker.GetCatl2 ( )

Gets catl2 value

Returns: catl2: L2 regularization in categorical split

◆ GetCatSmooth()

double Synapse.ML.Lightgbm.LightGBMRanker.GetCatSmooth ( )

Gets catSmooth value

Returns: catSmooth: this can reduce the effect of noises in categorical features, especially for categories with few data

◆ GetChunkSize()

int Synapse.ML.Lightgbm.LightGBMRanker.GetChunkSize ( )

Gets chunkSize value

Returns: chunkSize: Advanced parameter to specify the chunk size for copying Java data to native. If set too high, memory may be wasted, but if set too low, performance may be reduced during data copy.If dataset size is known beforehand, set to the number of rows in the dataset.

◆ GetDataRandomSeed()

int Synapse.ML.Lightgbm.LightGBMRanker.GetDataRandomSeed ( )

Gets dataRandomSeed value

Returns: dataRandomSeed: Random seed for sampling data to construct histogram bins.

◆ GetDataTransferMode()

string Synapse.ML.Lightgbm.LightGBMRanker.GetDataTransferMode ( )

Gets dataTransferMode value

Returns: dataTransferMode: Specify how SynapseML transfers data from Spark to LightGBM. Values can be streaming, bulk. Default is bulk, which is the legacy mode.

◆ GetDefaultListenPort()

int Synapse.ML.Lightgbm.LightGBMRanker.GetDefaultListenPort ( )

Gets defaultListenPort value

Returns: defaultListenPort: The default listen port on executors, used for testing

◆ GetDeterministic()

bool Synapse.ML.Lightgbm.LightGBMRanker.GetDeterministic ( )

Gets deterministic value

Returns: deterministic: Used only with cpu devide type. Setting this to true should ensure stable results when using the same data and the same parameters. Note: setting this to true may slow down training. To avoid potential instability due to numerical issues, please set force_col_wise=true or force_row_wise=true when setting deterministic=true

◆ GetDriverListenPort()

int Synapse.ML.Lightgbm.LightGBMRanker.GetDriverListenPort ( )

Gets driverListenPort value

Returns: driverListenPort: The listen port on a driver. Default value is 0 (random)

◆ GetDropRate()

double Synapse.ML.Lightgbm.LightGBMRanker.GetDropRate ( )

Gets dropRate value

Returns: dropRate: Dropout rate: a fraction of previous trees to drop during the dropout

◆ GetDropSeed()

int Synapse.ML.Lightgbm.LightGBMRanker.GetDropSeed ( )

Gets dropSeed value

Returns: dropSeed: Random seed to choose dropping models. Only used in dart.

◆ GetEarlyStoppingRound()

int Synapse.ML.Lightgbm.LightGBMRanker.GetEarlyStoppingRound ( )

Gets earlyStoppingRound value

Returns: earlyStoppingRound: Early stopping round

◆ GetEvalAt()

int [] Synapse.ML.Lightgbm.LightGBMRanker.GetEvalAt ( )

Gets evalAt value

Returns: evalAt: NDCG and MAP evaluation positions, separated by comma

◆ GetExecutionMode()

string Synapse.ML.Lightgbm.LightGBMRanker.GetExecutionMode ( )

Gets executionMode value

Returns: executionMode: Deprecated. Please use dataTransferMode.

◆ GetExtraSeed()

int Synapse.ML.Lightgbm.LightGBMRanker.GetExtraSeed ( )

Gets extraSeed value

Returns: extraSeed: Random seed for selecting threshold when extra_trees is true

◆ GetFeatureFraction()

double Synapse.ML.Lightgbm.LightGBMRanker.GetFeatureFraction ( )

Gets featureFraction value

Returns: featureFraction: Feature fraction

◆ GetFeatureFractionByNode()

double Synapse.ML.Lightgbm.LightGBMRanker.GetFeatureFractionByNode ( )

Gets featureFractionByNode value

Returns: featureFractionByNode: Feature fraction by node

◆ GetFeatureFractionSeed()

int Synapse.ML.Lightgbm.LightGBMRanker.GetFeatureFractionSeed ( )

Gets featureFractionSeed value

Returns: featureFractionSeed: Feature fraction seed

◆ GetFeaturesCol()

string Synapse.ML.Lightgbm.LightGBMRanker.GetFeaturesCol ( )

Gets featuresCol value

Returns: featuresCol: features column name

◆ GetFeaturesShapCol()

string Synapse.ML.Lightgbm.LightGBMRanker.GetFeaturesShapCol ( )

Gets featuresShapCol value

Returns: featuresShapCol: Output SHAP vector column name after prediction containing the feature contribution values

◆ GetFobj()

object Synapse.ML.Lightgbm.LightGBMRanker.GetFobj ( )

Gets fobj value

Returns: fobj: Customized objective function. Should accept two parameters: preds, train_data, and return (grad, hess).

◆ GetGroupCol()

string Synapse.ML.Lightgbm.LightGBMRanker.GetGroupCol ( )

Gets groupCol value

Returns: groupCol: The name of the group column

◆ GetImprovementTolerance()

double Synapse.ML.Lightgbm.LightGBMRanker.GetImprovementTolerance ( )

Gets improvementTolerance value

Returns: improvementTolerance: Tolerance to consider improvement in metric

◆ GetInitScoreCol()

string Synapse.ML.Lightgbm.LightGBMRanker.GetInitScoreCol ( )

Gets initScoreCol value

Returns: initScoreCol: The name of the initial score column, used for continued training

◆ GetIsEnableSparse()

bool Synapse.ML.Lightgbm.LightGBMRanker.GetIsEnableSparse ( )

Gets isEnableSparse value

Returns: isEnableSparse: Used to enable/disable sparse optimization

◆ GetIsProvideTrainingMetric()

bool Synapse.ML.Lightgbm.LightGBMRanker.GetIsProvideTrainingMetric ( )

Gets isProvideTrainingMetric value

Returns: isProvideTrainingMetric: Whether output metric result over training dataset.

◆ GetLabelCol()

string Synapse.ML.Lightgbm.LightGBMRanker.GetLabelCol ( )

Gets labelCol value

Returns: labelCol: label column name

◆ GetLabelGain()

double [] Synapse.ML.Lightgbm.LightGBMRanker.GetLabelGain ( )

Gets labelGain value

Returns: labelGain: graded relevance for each label in NDCG

◆ GetLambdaL1()

double Synapse.ML.Lightgbm.LightGBMRanker.GetLambdaL1 ( )

Gets lambdaL1 value

Returns: lambdaL1: L1 regularization

◆ GetLambdaL2()

double Synapse.ML.Lightgbm.LightGBMRanker.GetLambdaL2 ( )

Gets lambdaL2 value

Returns: lambdaL2: L2 regularization

◆ GetLeafPredictionCol()

string Synapse.ML.Lightgbm.LightGBMRanker.GetLeafPredictionCol ( )

Gets leafPredictionCol value

Returns: leafPredictionCol: Predicted leaf indices's column name

◆ GetLearningRate()

double Synapse.ML.Lightgbm.LightGBMRanker.GetLearningRate ( )

Gets learningRate value

Returns: learningRate: Learning rate or shrinkage rate

◆ GetMatrixType()

string Synapse.ML.Lightgbm.LightGBMRanker.GetMatrixType ( )

Gets matrixType value

Returns: matrixType: Advanced parameter to specify whether the native lightgbm matrix constructed should be sparse or dense. Values can be auto, sparse or dense. Default value is auto, which samples first ten rows to determine type.

◆ GetMaxBin()

int Synapse.ML.Lightgbm.LightGBMRanker.GetMaxBin ( )

Gets maxBin value

Returns: maxBin: Max bin

◆ GetMaxBinByFeature()

int [] Synapse.ML.Lightgbm.LightGBMRanker.GetMaxBinByFeature ( )

Gets maxBinByFeature value

Returns: maxBinByFeature: Max number of bins for each feature

◆ GetMaxCatThreshold()

int Synapse.ML.Lightgbm.LightGBMRanker.GetMaxCatThreshold ( )

Gets maxCatThreshold value

Returns: maxCatThreshold: limit number of split points considered for categorical features

◆ GetMaxCatToOnehot()

int Synapse.ML.Lightgbm.LightGBMRanker.GetMaxCatToOnehot ( )

Gets maxCatToOnehot value

Returns: maxCatToOnehot: when number of categories of one feature smaller than or equal to this, one-vs-other split algorithm will be used

◆ GetMaxDeltaStep()

double Synapse.ML.Lightgbm.LightGBMRanker.GetMaxDeltaStep ( )

Gets maxDeltaStep value

Returns: maxDeltaStep: Used to limit the max output of tree leaves

◆ GetMaxDepth()

int Synapse.ML.Lightgbm.LightGBMRanker.GetMaxDepth ( )

Gets maxDepth value

Returns: maxDepth: Max depth

◆ GetMaxDrop()

int Synapse.ML.Lightgbm.LightGBMRanker.GetMaxDrop ( )

Gets maxDrop value

Returns: maxDrop: Max number of dropped trees during one boosting iteration

◆ GetMaxNumClasses()

int Synapse.ML.Lightgbm.LightGBMRanker.GetMaxNumClasses ( )

Gets maxNumClasses value

Returns: maxNumClasses: Number of max classes to infer numClass in multi-class classification.

◆ GetMaxPosition()

int Synapse.ML.Lightgbm.LightGBMRanker.GetMaxPosition ( )

Gets maxPosition value

Returns: maxPosition: optimized NDCG at this position

◆ GetMaxStreamingOMPThreads()

int Synapse.ML.Lightgbm.LightGBMRanker.GetMaxStreamingOMPThreads ( )

Gets maxStreamingOMPThreads value

Returns: maxStreamingOMPThreads: Maximum number of OpenMP threads used by a LightGBM thread. Used only for thread-safe buffer allocation. Use -1 to use OpenMP default, but in a Spark environment it's best to set a fixed value.

◆ GetMetric()

string Synapse.ML.Lightgbm.LightGBMRanker.GetMetric ( )

Gets metric value

Returns: metric: Metrics to be evaluated on the evaluation data. Options are: empty string or not specified means that metric corresponding to specified objective will be used (this is possible only for pre-defined objective functions, otherwise no evaluation metric will be added). None (string, not a None value) means that no metric will be registered, aliases: na, null, custom. l1, absolute loss, aliases: mean_absolute_error, mae, regression_l1. l2, square loss, aliases: mean_squared_error, mse, regression_l2, regression. rmse, root square loss, aliases: root_mean_squared_error, l2_root. quantile, Quantile regression. mape, MAPE loss, aliases: mean_absolute_percentage_error. huber, Huber loss. fair, Fair loss. poisson, negative log-likelihood for Poisson regression. gamma, negative log-likelihood for Gamma regression. gamma_deviance, residual deviance for Gamma regression. tweedie, negative log-likelihood for Tweedie regression. ndcg, NDCG, aliases: lambdarank. map, MAP, aliases: mean_average_precision. auc, AUC. binary_logloss, log loss, aliases: binary. binary_error, for one sample: 0 for correct classification, 1 for error classification. multi_logloss, log loss for multi-class classification, aliases: multiclass, softmax, multiclassova, multiclass_ova, ova, ovr. multi_error, error rate for multi-class classification. cross_entropy, cross-entropy (with optional linear weights), aliases: xentropy. cross_entropy_lambda, intensity-weighted cross-entropy, aliases: xentlambda. kullback_leibler, Kullback-Leibler divergence, aliases: kldiv.

◆ GetMicroBatchSize()

int Synapse.ML.Lightgbm.LightGBMRanker.GetMicroBatchSize ( )

Gets microBatchSize value

Returns: microBatchSize: Specify how many elements are sent in a streaming micro-batch.

◆ GetMinDataInLeaf()

int Synapse.ML.Lightgbm.LightGBMRanker.GetMinDataInLeaf ( )

Gets minDataInLeaf value

Returns: minDataInLeaf: Minimal number of data in one leaf. Can be used to deal with over-fitting.

◆ GetMinDataPerBin()

int Synapse.ML.Lightgbm.LightGBMRanker.GetMinDataPerBin ( )

Gets minDataPerBin value

Returns: minDataPerBin: Minimal number of data inside one bin

◆ GetMinDataPerGroup()

int Synapse.ML.Lightgbm.LightGBMRanker.GetMinDataPerGroup ( )

Gets minDataPerGroup value

Returns: minDataPerGroup: minimal number of data per categorical group

◆ GetMinGainToSplit()

double Synapse.ML.Lightgbm.LightGBMRanker.GetMinGainToSplit ( )

Gets minGainToSplit value

Returns: minGainToSplit: The minimal gain to perform split

◆ GetMinSumHessianInLeaf()

double Synapse.ML.Lightgbm.LightGBMRanker.GetMinSumHessianInLeaf ( )

Gets minSumHessianInLeaf value

Returns: minSumHessianInLeaf: Minimal sum hessian in one leaf

◆ GetModelString()

string Synapse.ML.Lightgbm.LightGBMRanker.GetModelString ( )

Gets modelString value

Returns: modelString: LightGBM model to retrain

◆ GetMonotoneConstraints()

int [] Synapse.ML.Lightgbm.LightGBMRanker.GetMonotoneConstraints ( )

Gets monotoneConstraints value

Returns: monotoneConstraints: used for constraints of monotonic features. 1 means increasing, -1 means decreasing, 0 means non-constraint. Specify all features in order.

◆ GetMonotoneConstraintsMethod()

string Synapse.ML.Lightgbm.LightGBMRanker.GetMonotoneConstraintsMethod ( )

Gets monotoneConstraintsMethod value

Returns: monotoneConstraintsMethod: Monotone constraints method. basic, intermediate, or advanced.

◆ GetMonotonePenalty()

double Synapse.ML.Lightgbm.LightGBMRanker.GetMonotonePenalty ( )

Gets monotonePenalty value

Returns: monotonePenalty: A penalization parameter X forbids any monotone splits on the first X (rounded down) level(s) of the tree.

◆ GetNegBaggingFraction()

double Synapse.ML.Lightgbm.LightGBMRanker.GetNegBaggingFraction ( )

Gets negBaggingFraction value

Returns: negBaggingFraction: Negative Bagging fraction

◆ GetNumBatches()

int Synapse.ML.Lightgbm.LightGBMRanker.GetNumBatches ( )

Gets numBatches value

Returns: numBatches: If greater than 0, splits data into separate batches during training

◆ GetNumIterations()

int Synapse.ML.Lightgbm.LightGBMRanker.GetNumIterations ( )

Gets numIterations value

Returns: numIterations: Number of iterations, LightGBM constructs num_class * num_iterations trees

◆ GetNumLeaves()

int Synapse.ML.Lightgbm.LightGBMRanker.GetNumLeaves ( )

Gets numLeaves value

Returns: numLeaves: Number of leaves

◆ GetNumTasks()

int Synapse.ML.Lightgbm.LightGBMRanker.GetNumTasks ( )

Gets numTasks value

Returns: numTasks: Advanced parameter to specify the number of tasks. SynapseML tries to guess this based on cluster configuration, but this parameter can be used to override.

◆ GetNumThreads()

int Synapse.ML.Lightgbm.LightGBMRanker.GetNumThreads ( )

Gets numThreads value

Returns: numThreads: Number of threads per executor for LightGBM. For the best speed, set this to the number of real CPU cores.

◆ GetObjective()

string Synapse.ML.Lightgbm.LightGBMRanker.GetObjective ( )

Gets objective value

Returns: objective: The Objective. For regression applications, this can be: regression_l2, regression_l1, huber, fair, poisson, quantile, mape, gamma or tweedie. For classification applications, this can be: binary, multiclass, or multiclassova.

◆ GetObjectiveSeed()

int Synapse.ML.Lightgbm.LightGBMRanker.GetObjectiveSeed ( )

Gets objectiveSeed value

Returns: objectiveSeed: Random seed for objectives, if random process is needed. Currently used only for rank_xendcg objective.

◆ GetOtherRate()

double Synapse.ML.Lightgbm.LightGBMRanker.GetOtherRate ( )

Gets otherRate value

Returns: otherRate: The retain ratio of small gradient data. Only used in goss.

◆ GetParallelism()

string Synapse.ML.Lightgbm.LightGBMRanker.GetParallelism ( )

Gets parallelism value

Returns: parallelism: Tree learner parallelism, can be set to data_parallel or voting_parallel

◆ GetPassThroughArgs()

string Synapse.ML.Lightgbm.LightGBMRanker.GetPassThroughArgs ( )

Gets passThroughArgs value

Returns: passThroughArgs: Direct string to pass through to LightGBM library (appended with other explicitly set params). Will override any parameters given with explicit setters. Can include multiple parameters in one string. e.g., force_row_wise=true

◆ GetPosBaggingFraction()

double Synapse.ML.Lightgbm.LightGBMRanker.GetPosBaggingFraction ( )

Gets posBaggingFraction value

Returns: posBaggingFraction: Positive Bagging fraction

◆ GetPredictDisableShapeCheck()

bool Synapse.ML.Lightgbm.LightGBMRanker.GetPredictDisableShapeCheck ( )

Gets predictDisableShapeCheck value

Returns: predictDisableShapeCheck: control whether or not LightGBM raises an error when you try to predict on data with a different number of features than the training data

◆ GetPredictionCol()

string Synapse.ML.Lightgbm.LightGBMRanker.GetPredictionCol ( )

Gets predictionCol value

Returns: predictionCol: prediction column name

◆ GetReferenceDataset()

object Synapse.ML.Lightgbm.LightGBMRanker.GetReferenceDataset ( )

Gets referenceDataset value

Returns: referenceDataset: The reference Dataset that was used for the fit. If using samplingMode=custom, this must be set before fit().

◆ GetRepartitionByGroupingColumn()

bool Synapse.ML.Lightgbm.LightGBMRanker.GetRepartitionByGroupingColumn ( )

Gets repartitionByGroupingColumn value

Returns: repartitionByGroupingColumn: Repartition training data according to grouping column, on by default.

◆ GetSamplingMode()

string Synapse.ML.Lightgbm.LightGBMRanker.GetSamplingMode ( )

Gets samplingMode value

Returns: samplingMode: Data sampling for streaming mode. Sampled data is used to define bins. 'global': sample from all data, 'subset': sample from first N rows, or 'fixed': Take first N rows as sample.Values can be global, subset, or fixed. Default is subset.

◆ GetSamplingSubsetSize()

int Synapse.ML.Lightgbm.LightGBMRanker.GetSamplingSubsetSize ( )

Gets samplingSubsetSize value

Returns: samplingSubsetSize: Specify subset size N for the sampling mode 'subset'. 'binSampleCount' rows will be chosen from the first N values of the dataset. Subset can be used when rows are expected to be random and data is huge.

◆ GetSeed()

int Synapse.ML.Lightgbm.LightGBMRanker.GetSeed ( )

Gets seed value

Returns: seed: Main seed, used to generate other seeds

◆ GetSkipDrop()

double Synapse.ML.Lightgbm.LightGBMRanker.GetSkipDrop ( )

Gets skipDrop value

Returns: skipDrop: Probability of skipping the dropout procedure during a boosting iteration

◆ GetSlotNames()

string [] Synapse.ML.Lightgbm.LightGBMRanker.GetSlotNames ( )

Gets slotNames value

Returns: slotNames: List of slot names in the features column

◆ GetTimeout()

double Synapse.ML.Lightgbm.LightGBMRanker.GetTimeout ( )

Gets timeout value

Returns: timeout: Timeout in seconds

◆ GetTopK()

int Synapse.ML.Lightgbm.LightGBMRanker.GetTopK ( )

Gets topK value

Returns: topK: The top_k value used in Voting parallel, set this to larger value for more accurate result, but it will slow down the training speed. It should be greater than 0

◆ GetTopRate()

double Synapse.ML.Lightgbm.LightGBMRanker.GetTopRate ( )

Gets topRate value

Returns: topRate: The retain ratio of large gradient data. Only used in goss.

◆ GetUniformDrop()

bool Synapse.ML.Lightgbm.LightGBMRanker.GetUniformDrop ( )

Gets uniformDrop value

Returns: uniformDrop: Set this to true to use uniform drop in dart mode

◆ GetUseBarrierExecutionMode()

bool Synapse.ML.Lightgbm.LightGBMRanker.GetUseBarrierExecutionMode ( )

Gets useBarrierExecutionMode value

Returns: useBarrierExecutionMode: Barrier execution mode which uses a barrier stage, off by default.

◆ GetUseMissing()

bool Synapse.ML.Lightgbm.LightGBMRanker.GetUseMissing ( )

Gets useMissing value

Returns: useMissing: Set this to false to disable the special handle of missing value

◆ GetUseSingleDatasetMode()

bool Synapse.ML.Lightgbm.LightGBMRanker.GetUseSingleDatasetMode ( )

Gets useSingleDatasetMode value

Returns: useSingleDatasetMode: Use single dataset execution mode to create a single native dataset per executor (singleton) to reduce memory and communication overhead.

◆ GetValidationIndicatorCol()

string Synapse.ML.Lightgbm.LightGBMRanker.GetValidationIndicatorCol ( )

Gets validationIndicatorCol value

Returns: validationIndicatorCol: Indicates whether the row is for training or validation

◆ GetVerbosity()

int Synapse.ML.Lightgbm.LightGBMRanker.GetVerbosity ( )

Gets verbosity value

Returns: verbosity: Verbosity where lt 0 is Fatal, eq 0 is Error, eq 1 is Info, gt 1 is Debug

◆ GetWeightCol()

string Synapse.ML.Lightgbm.LightGBMRanker.GetWeightCol ( )

Gets weightCol value

Returns: weightCol: The name of the weight column

◆ GetXGBoostDartMode()

bool Synapse.ML.Lightgbm.LightGBMRanker.GetXGBoostDartMode ( )

Gets xGBoostDartMode value

Returns: xGBoostDartMode: Set this to true to use xgboost dart mode

◆ GetZeroAsMissing()

bool Synapse.ML.Lightgbm.LightGBMRanker.GetZeroAsMissing ( )

Gets zeroAsMissing value

Returns: zeroAsMissing: Set to true to treat all zero as missing values (including the unshown values in LibSVM / sparse matrices). Set to false to use na for representing missing values

◆ Load()

static LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.Load ( string path )

static

Loads the LightGBMRanker that was previously saved using Save(string).

Parameters

path	The path the previous LightGBMRanker was saved to

Returns: New LightGBMRanker object, loaded from path.

◆ Read()

JavaMLReader<LightGBMRanker> Synapse.ML.Lightgbm.LightGBMRanker.Read ( )

Get the corresponding JavaMLReader instance.

Returns: an JavaMLReader<LightGBMRanker> instance for this ML instance.

◆ Save()

void Synapse.ML.Lightgbm.LightGBMRanker.Save ( string path )

Saves the object so that it can be loaded later using Load. Note that these objects can be shared with Scala by Loading or Saving in Scala.

Parameters

path	The path to save the object to

◆ SetBaggingFraction()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetBaggingFraction ( double value )

Sets value for baggingFraction

Parameters

value Bagging fraction

Returns: New LightGBMRanker object

◆ SetBaggingFreq()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetBaggingFreq ( int value )

Sets value for baggingFreq

Parameters

value Bagging frequency

Returns: New LightGBMRanker object

◆ SetBaggingSeed()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetBaggingSeed ( int value )

Sets value for baggingSeed

Parameters

value Bagging seed

Returns: New LightGBMRanker object

◆ SetBinSampleCount()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetBinSampleCount ( int value )

Sets value for binSampleCount

Parameters

value Number of samples considered at computing histogram bins

Returns: New LightGBMRanker object

◆ SetBoostFromAverage()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetBoostFromAverage ( bool value )

Sets value for boostFromAverage

Parameters

value Adjusts initial score to the mean of labels for faster convergence

Returns: New LightGBMRanker object

◆ SetBoostingType()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetBoostingType ( string value )

Sets value for boostingType

Parameters

value Default gbdt = traditional Gradient Boosting Decision Tree. Options are: gbdt, gbrt, rf (Random Forest), random_forest, dart (Dropouts meet Multiple Additive Regression Trees), goss (Gradient-based One-Side Sampling).

Returns: New LightGBMRanker object

◆ SetCategoricalSlotIndexes()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetCategoricalSlotIndexes ( int[] value )

Sets value for categoricalSlotIndexes

Parameters

value List of categorical column indexes, the slot index in the features column

Returns: New LightGBMRanker object

◆ SetCategoricalSlotNames()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetCategoricalSlotNames ( string[] value )

Sets value for categoricalSlotNames

Parameters

value List of categorical column slot names, the slot name in the features column

Returns: New LightGBMRanker object

◆ SetCatl2()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetCatl2 ( double value )

Sets value for catl2

Parameters

value L2 regularization in categorical split

Returns: New LightGBMRanker object

◆ SetCatSmooth()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetCatSmooth ( double value )

Sets value for catSmooth

Parameters

value this can reduce the effect of noises in categorical features, especially for categories with few data

Returns: New LightGBMRanker object

◆ SetChunkSize()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetChunkSize ( int value )

Sets value for chunkSize

Parameters

value Advanced parameter to specify the chunk size for copying Java data to native. If set too high, memory may be wasted, but if set too low, performance may be reduced during data copy.If dataset size is known beforehand, set to the number of rows in the dataset.

Returns: New LightGBMRanker object

◆ SetDataRandomSeed()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetDataRandomSeed ( int value )

Sets value for dataRandomSeed

Parameters

value Random seed for sampling data to construct histogram bins.

Returns: New LightGBMRanker object

◆ SetDataTransferMode()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetDataTransferMode ( string value )

Sets value for dataTransferMode

Parameters

value Specify how SynapseML transfers data from Spark to LightGBM. Values can be streaming, bulk. Default is bulk, which is the legacy mode.

Returns: New LightGBMRanker object

◆ SetDefaultListenPort()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetDefaultListenPort ( int value )

Sets value for defaultListenPort

Parameters

value The default listen port on executors, used for testing

Returns: New LightGBMRanker object

◆ SetDeterministic()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetDeterministic ( bool value )

Sets value for deterministic

Parameters

value Used only with cpu devide type. Setting this to true should ensure stable results when using the same data and the same parameters. Note: setting this to true may slow down training. To avoid potential instability due to numerical issues, please set force_col_wise=true or force_row_wise=true when setting deterministic=true

Returns: New LightGBMRanker object

◆ SetDriverListenPort()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetDriverListenPort ( int value )

Sets value for driverListenPort

Parameters

value The listen port on a driver. Default value is 0 (random)

Returns: New LightGBMRanker object

◆ SetDropRate()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetDropRate ( double value )

Sets value for dropRate

Parameters

value Dropout rate: a fraction of previous trees to drop during the dropout

Returns: New LightGBMRanker object

◆ SetDropSeed()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetDropSeed ( int value )

Sets value for dropSeed

Parameters

value Random seed to choose dropping models. Only used in dart.

Returns: New LightGBMRanker object

◆ SetEarlyStoppingRound()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetEarlyStoppingRound ( int value )

Sets value for earlyStoppingRound

Parameters

value Early stopping round

Returns: New LightGBMRanker object

◆ SetEvalAt()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetEvalAt ( int[] value )

Sets value for evalAt

Parameters

value NDCG and MAP evaluation positions, separated by comma

Returns: New LightGBMRanker object

◆ SetExecutionMode()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetExecutionMode ( string value )

Sets value for executionMode

Parameters

value Deprecated. Please use dataTransferMode.

Returns: New LightGBMRanker object

◆ SetExtraSeed()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetExtraSeed ( int value )

Sets value for extraSeed

Parameters

value Random seed for selecting threshold when extra_trees is true

Returns: New LightGBMRanker object

◆ SetFeatureFraction()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetFeatureFraction ( double value )

Sets value for featureFraction

Parameters

value Feature fraction

Returns: New LightGBMRanker object

◆ SetFeatureFractionByNode()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetFeatureFractionByNode ( double value )

Sets value for featureFractionByNode

Parameters

value Feature fraction by node

Returns: New LightGBMRanker object

◆ SetFeatureFractionSeed()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetFeatureFractionSeed ( int value )

Sets value for featureFractionSeed

Parameters

value Feature fraction seed

Returns: New LightGBMRanker object

◆ SetFeaturesCol()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetFeaturesCol ( string value )

Sets value for featuresCol

Parameters

value features column name

Returns: New LightGBMRanker object

◆ SetFeaturesShapCol()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetFeaturesShapCol ( string value )

Sets value for featuresShapCol

Parameters

value Output SHAP vector column name after prediction containing the feature contribution values

Returns: New LightGBMRanker object

◆ SetFobj()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetFobj ( object value )

Sets value for fobj

Parameters

value Customized objective function. Should accept two parameters: preds, train_data, and return (grad, hess).

Returns: New LightGBMRanker object

◆ SetGroupCol()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetGroupCol ( string value )

Sets value for groupCol

Parameters

value The name of the group column

Returns: New LightGBMRanker object

◆ SetImprovementTolerance()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetImprovementTolerance ( double value )

Sets value for improvementTolerance

Parameters

value Tolerance to consider improvement in metric

Returns: New LightGBMRanker object

◆ SetInitScoreCol()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetInitScoreCol ( string value )

Sets value for initScoreCol

Parameters

value The name of the initial score column, used for continued training

Returns: New LightGBMRanker object

◆ SetIsEnableSparse()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetIsEnableSparse ( bool value )

Sets value for isEnableSparse

Parameters

value Used to enable/disable sparse optimization

Returns: New LightGBMRanker object

◆ SetIsProvideTrainingMetric()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetIsProvideTrainingMetric ( bool value )

Sets value for isProvideTrainingMetric

Parameters

value Whether output metric result over training dataset.

Returns: New LightGBMRanker object

◆ SetLabelCol()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetLabelCol ( string value )

Sets value for labelCol

Parameters

value label column name

Returns: New LightGBMRanker object

◆ SetLabelGain()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetLabelGain ( double[] value )

Sets value for labelGain

Parameters

value graded relevance for each label in NDCG

Returns: New LightGBMRanker object

◆ SetLambdaL1()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetLambdaL1 ( double value )

Sets value for lambdaL1

Parameters

value L1 regularization

Returns: New LightGBMRanker object

◆ SetLambdaL2()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetLambdaL2 ( double value )

Sets value for lambdaL2

Parameters

value L2 regularization

Returns: New LightGBMRanker object

◆ SetLeafPredictionCol()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetLeafPredictionCol ( string value )

Sets value for leafPredictionCol

Parameters

value Predicted leaf indices's column name

Returns: New LightGBMRanker object

◆ SetLearningRate()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetLearningRate ( double value )

Sets value for learningRate

Parameters

value Learning rate or shrinkage rate

Returns: New LightGBMRanker object

◆ SetMatrixType()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMatrixType ( string value )

Sets value for matrixType

Parameters

value Advanced parameter to specify whether the native lightgbm matrix constructed should be sparse or dense. Values can be auto, sparse or dense. Default value is auto, which samples first ten rows to determine type.

Returns: New LightGBMRanker object

◆ SetMaxBin()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMaxBin ( int value )

Sets value for maxBin

Parameters

value Max bin

Returns: New LightGBMRanker object

◆ SetMaxBinByFeature()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMaxBinByFeature ( int[] value )

Sets value for maxBinByFeature

Parameters

value Max number of bins for each feature

Returns: New LightGBMRanker object

◆ SetMaxCatThreshold()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMaxCatThreshold ( int value )

Sets value for maxCatThreshold

Parameters

value limit number of split points considered for categorical features

Returns: New LightGBMRanker object

◆ SetMaxCatToOnehot()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMaxCatToOnehot ( int value )

Sets value for maxCatToOnehot

Parameters

value when number of categories of one feature smaller than or equal to this, one-vs-other split algorithm will be used

Returns: New LightGBMRanker object

◆ SetMaxDeltaStep()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMaxDeltaStep ( double value )

Sets value for maxDeltaStep

Parameters

value Used to limit the max output of tree leaves

Returns: New LightGBMRanker object

◆ SetMaxDepth()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMaxDepth ( int value )

Sets value for maxDepth

Parameters

value Max depth

Returns: New LightGBMRanker object

◆ SetMaxDrop()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMaxDrop ( int value )

Sets value for maxDrop

Parameters

value Max number of dropped trees during one boosting iteration

Returns: New LightGBMRanker object

◆ SetMaxNumClasses()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMaxNumClasses ( int value )

Sets value for maxNumClasses

Parameters

value Number of max classes to infer numClass in multi-class classification.

Returns: New LightGBMRanker object

◆ SetMaxPosition()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMaxPosition ( int value )

Sets value for maxPosition

Parameters

value optimized NDCG at this position

Returns: New LightGBMRanker object

◆ SetMaxStreamingOMPThreads()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMaxStreamingOMPThreads ( int value )

Sets value for maxStreamingOMPThreads

Parameters

value Maximum number of OpenMP threads used by a LightGBM thread. Used only for thread-safe buffer allocation. Use -1 to use OpenMP default, but in a Spark environment it's best to set a fixed value.

Returns: New LightGBMRanker object

◆ SetMetric()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMetric ( string value )

Sets value for metric

Parameters

value Metrics to be evaluated on the evaluation data. Options are: empty string or not specified means that metric corresponding to specified objective will be used (this is possible only for pre-defined objective functions, otherwise no evaluation metric will be added). None (string, not a None value) means that no metric will be registered, aliases: na, null, custom. l1, absolute loss, aliases: mean_absolute_error, mae, regression_l1. l2, square loss, aliases: mean_squared_error, mse, regression_l2, regression. rmse, root square loss, aliases: root_mean_squared_error, l2_root. quantile, Quantile regression. mape, MAPE loss, aliases: mean_absolute_percentage_error. huber, Huber loss. fair, Fair loss. poisson, negative log-likelihood for Poisson regression. gamma, negative log-likelihood for Gamma regression. gamma_deviance, residual deviance for Gamma regression. tweedie, negative log-likelihood for Tweedie regression. ndcg, NDCG, aliases: lambdarank. map, MAP, aliases: mean_average_precision. auc, AUC. binary_logloss, log loss, aliases: binary. binary_error, for one sample: 0 for correct classification, 1 for error classification. multi_logloss, log loss for multi-class classification, aliases: multiclass, softmax, multiclassova, multiclass_ova, ova, ovr. multi_error, error rate for multi-class classification. cross_entropy, cross-entropy (with optional linear weights), aliases: xentropy. cross_entropy_lambda, intensity-weighted cross-entropy, aliases: xentlambda. kullback_leibler, Kullback-Leibler divergence, aliases: kldiv.

Returns: New LightGBMRanker object

◆ SetMicroBatchSize()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMicroBatchSize ( int value )

Sets value for microBatchSize

Parameters

value Specify how many elements are sent in a streaming micro-batch.

Returns: New LightGBMRanker object

◆ SetMinDataInLeaf()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMinDataInLeaf ( int value )

Sets value for minDataInLeaf

Parameters

value Minimal number of data in one leaf. Can be used to deal with over-fitting.

Returns: New LightGBMRanker object

◆ SetMinDataPerBin()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMinDataPerBin ( int value )

Sets value for minDataPerBin

Parameters

value Minimal number of data inside one bin

Returns: New LightGBMRanker object

◆ SetMinDataPerGroup()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMinDataPerGroup ( int value )

Sets value for minDataPerGroup

Parameters

value minimal number of data per categorical group

Returns: New LightGBMRanker object

◆ SetMinGainToSplit()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMinGainToSplit ( double value )

Sets value for minGainToSplit

Parameters

value The minimal gain to perform split

Returns: New LightGBMRanker object

◆ SetMinSumHessianInLeaf()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMinSumHessianInLeaf ( double value )

Sets value for minSumHessianInLeaf

Parameters

value Minimal sum hessian in one leaf

Returns: New LightGBMRanker object

◆ SetModelString()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetModelString ( string value )

Sets value for modelString

Parameters

value LightGBM model to retrain

Returns: New LightGBMRanker object

◆ SetMonotoneConstraints()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMonotoneConstraints ( int[] value )

Sets value for monotoneConstraints

Parameters

value used for constraints of monotonic features. 1 means increasing, -1 means decreasing, 0 means non-constraint. Specify all features in order.

Returns: New LightGBMRanker object

◆ SetMonotoneConstraintsMethod()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMonotoneConstraintsMethod ( string value )

Sets value for monotoneConstraintsMethod

Parameters

value Monotone constraints method. basic, intermediate, or advanced.

Returns: New LightGBMRanker object

◆ SetMonotonePenalty()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetMonotonePenalty ( double value )

Sets value for monotonePenalty

Parameters

value A penalization parameter X forbids any monotone splits on the first X (rounded down) level(s) of the tree.

Returns: New LightGBMRanker object

◆ SetNegBaggingFraction()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetNegBaggingFraction ( double value )

Sets value for negBaggingFraction

Parameters

value Negative Bagging fraction

Returns: New LightGBMRanker object

◆ SetNumBatches()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetNumBatches ( int value )

Sets value for numBatches

Parameters

value If greater than 0, splits data into separate batches during training

Returns: New LightGBMRanker object

◆ SetNumIterations()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetNumIterations ( int value )

Sets value for numIterations

Parameters

value Number of iterations, LightGBM constructs num_class * num_iterations trees

Returns: New LightGBMRanker object

◆ SetNumLeaves()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetNumLeaves ( int value )

Sets value for numLeaves

Parameters

value Number of leaves

Returns: New LightGBMRanker object

◆ SetNumTasks()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetNumTasks ( int value )

Sets value for numTasks

Parameters

value Advanced parameter to specify the number of tasks. SynapseML tries to guess this based on cluster configuration, but this parameter can be used to override.

Returns: New LightGBMRanker object

◆ SetNumThreads()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetNumThreads ( int value )

Sets value for numThreads

Parameters

value Number of threads per executor for LightGBM. For the best speed, set this to the number of real CPU cores.

Returns: New LightGBMRanker object

◆ SetObjective()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetObjective ( string value )

Sets value for objective

Parameters

value The Objective. For regression applications, this can be: regression_l2, regression_l1, huber, fair, poisson, quantile, mape, gamma or tweedie. For classification applications, this can be: binary, multiclass, or multiclassova.

Returns: New LightGBMRanker object

◆ SetObjectiveSeed()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetObjectiveSeed ( int value )

Sets value for objectiveSeed

Parameters

value Random seed for objectives, if random process is needed. Currently used only for rank_xendcg objective.

Returns: New LightGBMRanker object

◆ SetOtherRate()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetOtherRate ( double value )

Sets value for otherRate

Parameters

value The retain ratio of small gradient data. Only used in goss.

Returns: New LightGBMRanker object

◆ SetParallelism()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetParallelism ( string value )

Sets value for parallelism

Parameters

value Tree learner parallelism, can be set to data_parallel or voting_parallel

Returns: New LightGBMRanker object

◆ SetPassThroughArgs()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetPassThroughArgs ( string value )

Sets value for passThroughArgs

Parameters

value Direct string to pass through to LightGBM library (appended with other explicitly set params). Will override any parameters given with explicit setters. Can include multiple parameters in one string. e.g., force_row_wise=true

Returns: New LightGBMRanker object

◆ SetPosBaggingFraction()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetPosBaggingFraction ( double value )

Sets value for posBaggingFraction

Parameters

value Positive Bagging fraction

Returns: New LightGBMRanker object

◆ SetPredictDisableShapeCheck()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetPredictDisableShapeCheck ( bool value )

Sets value for predictDisableShapeCheck

Parameters

value control whether or not LightGBM raises an error when you try to predict on data with a different number of features than the training data

Returns: New LightGBMRanker object

◆ SetPredictionCol()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetPredictionCol ( string value )

Sets value for predictionCol

Parameters

value prediction column name

Returns: New LightGBMRanker object

◆ SetReferenceDataset()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetReferenceDataset ( object value )

Sets value for referenceDataset

Parameters

value The reference Dataset that was used for the fit. If using samplingMode=custom, this must be set before fit().

Returns: New LightGBMRanker object

◆ SetRepartitionByGroupingColumn()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetRepartitionByGroupingColumn ( bool value )

Sets value for repartitionByGroupingColumn

Parameters

value Repartition training data according to grouping column, on by default.

Returns: New LightGBMRanker object

◆ SetSamplingMode()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetSamplingMode ( string value )

Sets value for samplingMode

Parameters

value Data sampling for streaming mode. Sampled data is used to define bins. 'global': sample from all data, 'subset': sample from first N rows, or 'fixed': Take first N rows as sample.Values can be global, subset, or fixed. Default is subset.

Returns: New LightGBMRanker object

◆ SetSamplingSubsetSize()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetSamplingSubsetSize ( int value )

Sets value for samplingSubsetSize

Parameters

value Specify subset size N for the sampling mode 'subset'. 'binSampleCount' rows will be chosen from the first N values of the dataset. Subset can be used when rows are expected to be random and data is huge.

Returns: New LightGBMRanker object

◆ SetSeed()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetSeed ( int value )

Sets value for seed

Parameters

value Main seed, used to generate other seeds

Returns: New LightGBMRanker object

◆ SetSkipDrop()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetSkipDrop ( double value )

Sets value for skipDrop

Parameters

value Probability of skipping the dropout procedure during a boosting iteration

Returns: New LightGBMRanker object

◆ SetSlotNames()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetSlotNames ( string[] value )

Sets value for slotNames

Parameters

value List of slot names in the features column

Returns: New LightGBMRanker object

◆ SetTimeout()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetTimeout ( double value )

Sets value for timeout

Parameters

value Timeout in seconds

Returns: New LightGBMRanker object

◆ SetTopK()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetTopK ( int value )

Sets value for topK

Parameters

value The top_k value used in Voting parallel, set this to larger value for more accurate result, but it will slow down the training speed. It should be greater than 0

Returns: New LightGBMRanker object

◆ SetTopRate()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetTopRate ( double value )

Sets value for topRate

Parameters

value The retain ratio of large gradient data. Only used in goss.

Returns: New LightGBMRanker object

◆ SetUniformDrop()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetUniformDrop ( bool value )

Sets value for uniformDrop

Parameters

value Set this to true to use uniform drop in dart mode

Returns: New LightGBMRanker object

◆ SetUseBarrierExecutionMode()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetUseBarrierExecutionMode ( bool value )

Sets value for useBarrierExecutionMode

Parameters

value Barrier execution mode which uses a barrier stage, off by default.

Returns: New LightGBMRanker object

◆ SetUseMissing()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetUseMissing ( bool value )

Sets value for useMissing

Parameters

value Set this to false to disable the special handle of missing value

Returns: New LightGBMRanker object

◆ SetUseSingleDatasetMode()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetUseSingleDatasetMode ( bool value )

Sets value for useSingleDatasetMode

Parameters

value Use single dataset execution mode to create a single native dataset per executor (singleton) to reduce memory and communication overhead.

Returns: New LightGBMRanker object

◆ SetValidationIndicatorCol()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetValidationIndicatorCol ( string value )

Sets value for validationIndicatorCol

Parameters

value Indicates whether the row is for training or validation

Returns: New LightGBMRanker object

◆ SetVerbosity()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetVerbosity ( int value )

Sets value for verbosity

Parameters

value Verbosity where lt 0 is Fatal, eq 0 is Error, eq 1 is Info, gt 1 is Debug

Returns: New LightGBMRanker object

◆ SetWeightCol()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetWeightCol ( string value )

Sets value for weightCol

Parameters

value The name of the weight column

Returns: New LightGBMRanker object

◆ SetXGBoostDartMode()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetXGBoostDartMode ( bool value )

Sets value for xGBoostDartMode

Parameters

value Set this to true to use xgboost dart mode

Returns: New LightGBMRanker object

◆ SetZeroAsMissing()

LightGBMRanker Synapse.ML.Lightgbm.LightGBMRanker.SetZeroAsMissing ( bool value )

Sets value for zeroAsMissing

Parameters

value Set to true to treat all zero as missing values (including the unshown values in LibSVM / sparse matrices). Set to false to use na for representing missing values

Returns: New LightGBMRanker object

The documentation for this class was generated from the following file:

synapse/ml/lightgbm/LightGBMRanker.cs

Public Member Functions

Static Public Member Functions

Detailed Description

Constructor & Destructor Documentation

◆ LightGBMRanker() [1/2]

◆ LightGBMRanker() [2/2]

Member Function Documentation

◆ Fit()

◆ GetBaggingFraction()

◆ GetBaggingFreq()

◆ GetBaggingSeed()

◆ GetBinSampleCount()

◆ GetBoostFromAverage()

◆ GetBoostingType()

◆ GetCategoricalSlotIndexes()

◆ GetCategoricalSlotNames()

◆ GetCatl2()

◆ GetCatSmooth()

◆ GetChunkSize()

◆ GetDataRandomSeed()

◆ GetDataTransferMode()

◆ GetDefaultListenPort()

◆ GetDeterministic()

◆ GetDriverListenPort()

◆ GetDropRate()

◆ GetDropSeed()

◆ GetEarlyStoppingRound()

◆ GetEvalAt()

◆ GetExecutionMode()

◆ GetExtraSeed()

◆ GetFeatureFraction()

◆ GetFeatureFractionByNode()

◆ GetFeatureFractionSeed()

◆ GetFeaturesCol()

◆ GetFeaturesShapCol()

◆ GetFobj()

◆ GetGroupCol()

◆ GetImprovementTolerance()

◆ GetInitScoreCol()

◆ GetIsEnableSparse()

◆ GetIsProvideTrainingMetric()

◆ GetLabelCol()

◆ GetLabelGain()

◆ GetLambdaL1()

◆ GetLambdaL2()

◆ GetLeafPredictionCol()

◆ GetLearningRate()

◆ GetMatrixType()

◆ GetMaxBin()

◆ GetMaxBinByFeature()

◆ GetMaxCatThreshold()

◆ GetMaxCatToOnehot()

◆ GetMaxDeltaStep()

◆ GetMaxDepth()

◆ GetMaxDrop()

◆ GetMaxNumClasses()

◆ GetMaxPosition()

◆ GetMaxStreamingOMPThreads()

◆ GetMetric()

◆ GetMicroBatchSize()

◆ GetMinDataInLeaf()

◆ GetMinDataPerBin()

◆ GetMinDataPerGroup()

◆ GetMinGainToSplit()

◆ GetMinSumHessianInLeaf()

◆ GetModelString()

◆ GetMonotoneConstraints()

◆ GetMonotoneConstraintsMethod()

◆ GetMonotonePenalty()

◆ GetNegBaggingFraction()

◆ GetNumBatches()

◆ GetNumIterations()

◆ GetNumLeaves()

◆ GetNumTasks()

◆ GetNumThreads()

◆ GetObjective()

◆ GetObjectiveSeed()

◆ GetOtherRate()

◆ GetParallelism()

◆ GetPassThroughArgs()