package lightgbm
- Alphabetic
- Public
- All
Type Members
-
abstract
class
BasePartitionTask extends Serializable with Logging
Class for handling the execution of Tasks on workers for each partition.
Class for handling the execution of Tasks on workers for each partition. Only runs on worker Tasks.
-
class
BulkPartitionTask extends BasePartitionTask
Class for handling the execution of bulk-based Tasks on workers for each partition.
- case class ColumnParams(labelColumn: String, featuresColumn: String, weightColumn: Option[String], initScoreColumn: Option[String], groupColumn: Option[String]) extends Product with Serializable
-
class
GroupIdManager extends AnyRef
Class for converting column values to group ID.
Class for converting column values to group ID.
Ints can just be returned, but a map of Long and String values is maintained so that unique and consistent values can be returned.
-
trait
HasActualNumClasses extends Params
Special parameter for classification model for actual number of classes in dataset
-
class
InstrumentationMeasures extends Serializable
Class for encapsulating performance instrumentation measures of overall training.
- trait LightGBMBase[TrainedModel <: Model[TrainedModel] with LightGBMModelParams] extends Estimator[TrainedModel] with LightGBMParams with ComplexParamsWritable with HasFeaturesCol with HasLabelCol with LightGBMPerformance with SynapseMLLogging
-
class
LightGBMClassificationModel extends ProbabilisticClassificationModel[Vector, LightGBMClassificationModel] with LightGBMModelParams with LightGBMModelMethods with LightGBMPredictionParams with HasActualNumClasses with ComplexParamsWritable with SynapseMLLogging
Model produced by LightGBMClassifier.
-
class
LightGBMClassifier extends ProbabilisticClassifier[Vector, LightGBMClassifier, LightGBMClassificationModel] with LightGBMBase[LightGBMClassificationModel] with SynapseMLLogging
Trains a LightGBM Classification model, a fast, distributed, high performance gradient boosting framework based on decision tree algorithms.
Trains a LightGBM Classification model, a fast, distributed, high performance gradient boosting framework based on decision tree algorithms. For more information please see here: https://github.com/Microsoft/LightGBM. For parameter information see here: https://github.com/Microsoft/LightGBM/blob/master/docs/Parameters.rst
- trait LightGBMDelegate extends Serializable
-
trait
LightGBMModelMethods extends LightGBMModelParams with Logging
Contains common LightGBM model methods across all LightGBM learner types.
- trait LightGBMPerformance extends Serializable
-
class
LightGBMRanker extends Ranker[Vector, LightGBMRanker, LightGBMRankerModel] with LightGBMBase[LightGBMRankerModel] with SynapseMLLogging
Trains a LightGBMRanker model, a fast, distributed, high performance gradient boosting framework based on decision tree algorithms.
Trains a LightGBMRanker model, a fast, distributed, high performance gradient boosting framework based on decision tree algorithms. For more information please see here: https://github.com/Microsoft/LightGBM. For parameter information see here: https://github.com/Microsoft/LightGBM/blob/master/docs/Parameters.rst
-
class
LightGBMRankerModel extends RankerModel[Vector, LightGBMRankerModel] with LightGBMModelParams with LightGBMModelMethods with LightGBMPredictionParams with ComplexParamsWritable with SynapseMLLogging
Model produced by LightGBMRanker.
-
class
LightGBMRegressionModel extends RegressionModel[Vector, LightGBMRegressionModel] with LightGBMModelParams with LightGBMModelMethods with LightGBMPredictionParams with ComplexParamsWritable with SynapseMLLogging
Model produced by LightGBMRegressor.
-
class
LightGBMRegressor extends BaseRegressor[Vector, LightGBMRegressor, LightGBMRegressionModel] with LightGBMBase[LightGBMRegressionModel] with SynapseMLLogging
Trains a LightGBM Regression model, a fast, distributed, high performance gradient boosting framework based on decision tree algorithms.
Trains a LightGBM Regression model, a fast, distributed, high performance gradient boosting framework based on decision tree algorithms. For more information please see here: https://github.com/Microsoft/LightGBM. For parameter information see here: https://github.com/Microsoft/LightGBM/blob/master/docs/Parameters.rst Note: The application parameter supports the following values:
- regression_l2, L2 loss, alias=regression, mean_squared_error, mse, l2_root, root_mean_squared_error, rmse
- regression_l1, L1 loss, alias=mean_absolute_error, mae
- huber, Huber loss
- fair, Fair loss
- poisson, Poisson regression
- quantile, Quantile regression
- mape, MAPE loss, alias=mean_absolute_percentage_error
- gamma, Gamma regression with log-link. It might be useful, e.g., for modeling insurance claims severity, or for any target that might be gamma-distributed
- tweedie, Tweedie regression with log-link. It might be useful, e.g., for modeling total loss in insurance, or for any target that might be tweedie-distributed
-
case class
NetworkManager(numTasks: Int, driverServerSocket: ServerSocket, host: String, port: Int, timeout: Double, useBarrierExecutionMode: Boolean) extends Logging with Product with Serializable
Object to encapsulate all Spark/LightGBM network topology information, along with operations on the network.
- case class NetworkParams(defaultListenPort: Int, ipAddress: String, port: Int, barrierExecutionMode: Boolean) extends Product with Serializable
- case class NetworkTopologyInfo(lightgbmNetworkString: String, executorPartitionIdList: Array[Int], localListenPort: Int) extends Product with Serializable
-
case class
PartitionDataState(aggregatedTrainingData: Option[BaseAggregatedColumns], aggregatedValidationData: Option[BaseAggregatedColumns]) extends Product with Serializable
Object to encapsulate all intermediate data calculations.
Object to encapsulate all intermediate data calculations. Note tha only bulk mode uses these properties, but BasePartitionTask uses this class for consistent interfaces.
-
case class
PartitionResult(booster: Option[LightGBMBooster], taskMeasures: TaskInstrumentationMeasures) extends Product with Serializable
Object to encapsulate results from mapPartitions call.
-
case class
PartitionTaskContext(trainingCtx: TrainingContext, partitionId: Int, taskId: Long, measures: TaskInstrumentationMeasures, networkTopologyInfo: NetworkTopologyInfo, shouldExecuteTraining: Boolean, isEmptyPartition: Boolean, shouldReturnBooster: Boolean, shouldCalcValidationDataset: Boolean) extends Product with Serializable
Object to encapsulate most setup information about a particular partition Task
-
case class
PartitionTaskTrainingState(ctx: PartitionTaskContext, booster: LightGBMBooster) extends Product with Serializable
Object to encapsulate all training state on a single partition, plus the actual Booster
- class SharedDatasetState extends AnyRef
- class SharedState extends AnyRef
-
class
StreamingPartitionTask extends BasePartitionTask
Class for handling the execution of streaming-based Tasks on workers for each partition.
- case class StreamingState(ctx: PartitionTaskContext, dataset: LightGBMDataset, threadIndex: Int) extends Product with Serializable
-
class
TaskInstrumentationMeasures extends Serializable
Class for encapsulating performance instrumentation measures of each partition Task.
- case class TaskMessageInfo(status: String, taskHost: String, localListenPort: Int, partitionId: Int, executorId: String) extends Product with Serializable
-
case class
TrainingContext(batchIndex: Int, sharedStateSingleton: SharedSingleton[SharedState], schema: StructType, numCols: Int, numInitScoreClasses: Int, trainingParams: BaseTrainParams, networkParams: NetworkParams, columnParams: ColumnParams, datasetParams: String, featureNames: Option[Array[String]], numTasksPerExecutor: Int, validationData: Option[Broadcast[Array[Row]]], serializedReferenceDataset: Option[Array[Byte]], partitionCounts: Option[Array[Long]]) extends Serializable with Product
Object to encapsulate all information about a training session that does not change during execution and can be created on the driver.
Object to encapsulate all information about a training session that does not change during execution and can be created on the driver. There is also a reference to the shared state in an executor, which can change over time.
Value Members
- object InstrumentationMeasures extends Serializable
- object LightGBMClassificationModel extends ComplexParamsReadable[LightGBMClassificationModel] with Serializable
- object LightGBMClassifier extends DefaultParamsReadable[LightGBMClassifier] with Serializable
- object LightGBMConstants
- object LightGBMRanker extends DefaultParamsReadable[LightGBMRanker] with Serializable
- object LightGBMRankerModel extends ComplexParamsReadable[LightGBMRankerModel] with Serializable
- object LightGBMRegressionModel extends ComplexParamsReadable[LightGBMRegressionModel] with Serializable
- object LightGBMRegressor extends DefaultParamsReadable[LightGBMRegressor] with Serializable
-
object
LightGBMUtils
Helper utilities for LightGBM learners
- object NetworkManager extends Serializable