File predictor_probability_marginalized.hpp¶
-
namespace boosting¶
-
class IMarginalizedProbabilityPredictorConfig¶
- #include <predictor_probability_marginalized.hpp>
Defines an interface for all classes that allow to configure a predictor that predicts label-wise probabilities for given query examples by marginalizing over the joint probabilities of known label vectors.
Subclassed by boosting::MarginalizedProbabilityPredictorConfig
Public Functions
-
inline virtual ~IMarginalizedProbabilityPredictorConfig()¶
-
virtual bool isProbabilityCalibrationModelUsed() const = 0¶
Returns whether a model for the calibration of probabilities is used, if available, or not.
- Returns:
True, if a model for the calibration of probabilities is used, if available, false otherwise
-
virtual IMarginalizedProbabilityPredictorConfig &setUseProbabilityCalibrationModel(bool useProbabilityCalibrationModel) = 0¶
Sets whether a model for the calibration of probabilities should be used, if available, or not.
- Parameters:
useProbabilityCalibrationModel – True, if a model for the calibration of probabilities should be used, if available, false otherwise
- Returns:
A reference to an object of type
IMarginalizedProbabilityPredictorConfig
that allows further configuration of the predictor
-
inline virtual ~IMarginalizedProbabilityPredictorConfig()¶
-
class MarginalizedProbabilityPredictorConfig : public boosting::IMarginalizedProbabilityPredictorConfig, public IProbabilityPredictorConfig¶
- #include <predictor_probability_marginalized.hpp>
Allows to configure a predictor that predicts label-wise probabilities for given query examples by marginalizing over the joint probabilities of known label vectors.
Public Functions
-
MarginalizedProbabilityPredictorConfig(const std::unique_ptr<ILossConfig> &lossConfigPtr, const std::unique_ptr<IMultiThreadingConfig> &multiThreadingConfigPtr)¶
- Parameters:
lossConfigPtr – A reference to an unique pointer that stores the configuration of the loss function
multiThreadingConfigPtr – A reference to an unique pointer that stores the configuration of the multi-threading behavior that should be used to predict for several query examples in parallel
-
virtual bool isProbabilityCalibrationModelUsed() const override¶
Returns whether a model for the calibration of probabilities is used, if available, or not.
- Returns:
True, if a model for the calibration of probabilities is used, if available, false otherwise
-
virtual IMarginalizedProbabilityPredictorConfig &setUseProbabilityCalibrationModel(bool useProbabilityCalibrationModel) override¶
Sets whether a model for the calibration of probabilities should be used, if available, or not.
- Parameters:
useProbabilityCalibrationModel – True, if a model for the calibration of probabilities should be used, if available, false otherwise
- Returns:
A reference to an object of type
IMarginalizedProbabilityPredictorConfig
that allows further configuration of the predictor
-
std::unique_ptr<IProbabilityPredictorFactory> createPredictorFactory(const IRowWiseFeatureMatrix &featureMatrix, uint32 numLabels) const override¶
See also
IProbabilityPredictorConfig::createPredictorFactory
-
bool isLabelVectorSetNeeded() const override¶
See also
IPredictorConfig::isLabelVectorSetNeeded
Private Members
-
std::unique_ptr<IMarginalProbabilityCalibrationModel> noMarginalProbabilityCalibrationModelPtr_¶
-
std::unique_ptr<IJointProbabilityCalibrationModel> noJointProbabilityCalibrationModelPtr_¶
-
const std::unique_ptr<ILossConfig> &lossConfigPtr_¶
-
const std::unique_ptr<IMultiThreadingConfig> &multiThreadingConfigPtr_¶
-
MarginalizedProbabilityPredictorConfig(const std::unique_ptr<ILossConfig> &lossConfigPtr, const std::unique_ptr<IMultiThreadingConfig> &multiThreadingConfigPtr)¶
-
class IMarginalizedProbabilityPredictorConfig¶