mlrl.boosting.config module¶
Author: Michael Rapp (michael.rapp.ml@gmail.com)
Provides utility function for configuring boosting algorithms.
- class mlrl.boosting.config.BinaryPredictorParameter¶
Bases:
NominalParameter
A parameter that allows to configure the strategy to be used for predicting binary labels.
- BINARY_PREDICTOR_EXAMPLE_WISE = 'example-wise'¶
- BINARY_PREDICTOR_GFM = 'gfm'¶
- BINARY_PREDICTOR_LABEL_WISE = 'label-wise'¶
- class mlrl.boosting.config.DefaultRuleParameter¶
Bases:
NominalParameter
A parameter that allows to configure whether a default rule should be induced or not.
- class mlrl.boosting.config.ExtendedFeatureBinningParameter¶
Bases:
FeatureBinningParameter
Extends the FeatureBinningParameter by a value for automatic configuration.
- class mlrl.boosting.config.ExtendedParallelRuleRefinementParameter¶
Bases:
ParallelRuleRefinementParameter
Extends the ParallelRuleRefinementParameter by a value for automatic configuration.
- class mlrl.boosting.config.ExtendedParallelStatisticUpdateParameter¶
Bases:
ParallelStatisticUpdateParameter
Extends the ParallelStatisticUpdateParameter by a value for automatic configuration.
- class mlrl.boosting.config.ExtendedPartitionSamplingParameter¶
Bases:
PartitionSamplingParameter
Extends the PartitionSamplingParameter by a value for automatic configuration.
- class mlrl.boosting.config.HeadTypeParameter¶
Bases:
NominalParameter
A parameter that allows to configure the type of the rule heads that should be used.
- HEAD_TYPE_COMPLETE = 'complete'¶
- HEAD_TYPE_PARTIAL_DYNAMIC = 'partial-dynamic'¶
- HEAD_TYPE_PARTIAL_FIXED = 'partial-fixed'¶
- HEAD_TYPE_SINGLE = 'single-label'¶
- OPTION_EXPONENT = 'exponent'¶
- OPTION_LABEL_RATIO = 'label_ratio'¶
- OPTION_MAX_LABELS = 'max_labels'¶
- OPTION_MIN_LABELS = 'min_labels'¶
- OPTION_THRESHOLD = 'threshold'¶
- class mlrl.boosting.config.JointProbabilityCalibrationParameter¶
Bases:
NominalParameter
A parameter that allows to configure the method to be used for the calibration of joint probabilities.
- class mlrl.boosting.config.L1RegularizationParameter¶
Bases:
FloatParameter
A parameter that allows to configure the weight of the L1 regularization.
- class mlrl.boosting.config.L2RegularizationParameter¶
Bases:
FloatParameter
A parameter that allows to configure the weight of the L2 regularization.
- class mlrl.boosting.config.LabelBinningParameter¶
Bases:
NominalParameter
A parameter that allows to configure the strategy to be used for gradient-based label binning (GBLB).
- class mlrl.boosting.config.LossParameter¶
Bases:
NominalParameter
A parameter that allows to configure the loss function to be minimized during training.
- LOSS_LOGISTIC_EXAMPLE_WISE = 'logistic-example-wise'¶
- LOSS_LOGISTIC_LABEL_WISE = 'logistic-label-wise'¶
- LOSS_SQUARED_ERROR_EXAMPLE_WISE = 'squared-error-example-wise'¶
- LOSS_SQUARED_ERROR_LABEL_WISE = 'squared-error-label-wise'¶
- LOSS_SQUARED_HINGE_EXAMPLE_WISE = 'squared-hinge-example-wise'¶
- LOSS_SQUARED_HINGE_LABEL_WISE = 'squared-hinge-label-wise'¶
- class mlrl.boosting.config.MarginalProbabilityCalibrationParameter¶
Bases:
NominalParameter
A parameter that allows to configure the method to be used for the calibration of marginal probabilities.
- class mlrl.boosting.config.ProbabilityPredictorParameter¶
Bases:
NominalParameter
A parameter that allows to configure the strategy to be used for predicting probabilities.
- PROBABILITY_PREDICTOR_LABEL_WISE = 'label-wise'¶
- PROBABILITY_PREDICTOR_MARGINALIZED = 'marginalized'¶
- class mlrl.boosting.config.ShrinkageParameter¶
Bases:
FloatParameter
A parameter that allows to configure the shrinkage parameter, a.k.a. the learning rate, to be used.
- class mlrl.boosting.config.StatisticFormatParameter¶
Bases:
NominalParameter
A parameter that allows to configure the format to be used for the representation of gradients and Hessians.
- STATISTIC_FORMAT_DENSE = 'dense'¶
- STATISTIC_FORMAT_SPARSE = 'sparse'¶