Packages

package lightgbm

Ordering
  1. Alphabetic
Visibility
  1. Public
  2. All

Type Members

  1. case class ColumnParams(labelColumn: String, featuresColumn: String, weightColumn: Option[String], initScoreColumn: Option[String], groupColumn: Option[String]) extends Product with Serializable
  2. trait HasActualNumClasses extends Params

    Special parameter for classification model for actual number of classes in dataset

  3. trait LightGBMBase[TrainedModel <: Model[TrainedModel]] extends Estimator[TrainedModel] with LightGBMParams with HasFeaturesCol with HasLabelCol with BasicLogging
  4. class LightGBMClassificationModel extends ProbabilisticClassificationModel[Vector, LightGBMClassificationModel] with LightGBMModelParams with LightGBMModelMethods with LightGBMPredictionParams with HasActualNumClasses with ComplexParamsWritable with BasicLogging

    Model produced by LightGBMClassifier.

  5. class LightGBMClassifier extends ProbabilisticClassifier[Vector, LightGBMClassifier, LightGBMClassificationModel] with LightGBMBase[LightGBMClassificationModel] with BasicLogging

    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

  6. trait LightGBMDelegate extends Serializable
  7. trait LightGBMModelMethods extends LightGBMModelParams with Logging

    Contains common LightGBM model methods across all LightGBM learner types.

  8. class LightGBMRanker extends Ranker[Vector, LightGBMRanker, LightGBMRankerModel] with LightGBMBase[LightGBMRankerModel] with BasicLogging

    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

  9. class LightGBMRankerModel extends RankerModel[Vector, LightGBMRankerModel] with LightGBMModelParams with LightGBMModelMethods with LightGBMPredictionParams with ComplexParamsWritable with BasicLogging

    Model produced by LightGBMRanker.

  10. class LightGBMRegressionModel extends RegressionModel[Vector, LightGBMRegressionModel] with LightGBMModelParams with LightGBMModelMethods with LightGBMPredictionParams with ComplexParamsWritable with BasicLogging

    Model produced by LightGBMRegressor.

  11. class LightGBMRegressor extends BaseRegressor[Vector, LightGBMRegressor, LightGBMRegressionModel] with LightGBMBase[LightGBMRegressionModel] with BasicLogging

    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
  12. case class NetworkParams(defaultListenPort: Int, addr: String, port: Int, barrierExecutionMode: Boolean) extends Product with Serializable
  13. class SharedDatasetState extends AnyRef
  14. class SharedState extends AnyRef

Ungrouped