mlrl.boosting.testbed.experiments.output.probability_calibration.writer module¶

Author: Michael Rapp (michael.rapp.ml@gmail.com)

Provides classes that allow writing textual representations of probability calibration models to one or several sinks.

class mlrl.boosting.testbed.experiments.output.probability_calibration.writer.JointProbabilityCalibrationModelWriter(*extractors: DataExtractor)¶

Bases: ProbabilityCalibrationModelWriter

Allows writing textual representations of models for the calibration of joint probabilities to one or several sinks.

CONTEXT = Context(include_dataset_type=False, include_prediction_scope=True, include_fold=True)¶

class DefaultExtractor¶

Bases: DefaultExtractor

Extracts isotonic regression models for the calibration of joint probabilities that are stores as part of a rule model.

class InputExtractor(properties: TabularProperties, context: Context)¶

Bases: InputExtractor

Uses TabularInputData that has previously been loaded via an input reader.

PROPERTIES = TabularProperties(name='Joint probability calibration model', file_name='joint_probability_calibration_model', has_header=True)¶

class mlrl.boosting.testbed.experiments.output.probability_calibration.writer.MarginalProbabilityCalibrationModelWriter(*extractors: DataExtractor)¶

Bases: ProbabilityCalibrationModelWriter

Allows writing textual representations of models for the calibration of marginal probabilities to one or several sinks.

CONTEXT = Context(include_dataset_type=False, include_prediction_scope=True, include_fold=True)¶

class DefaultExtractor¶

Bases: DefaultExtractor

Extracts isotonic regression models for the calibration of marginal probabilities that are stores as part of a rule model.

class InputExtractor(properties: TabularProperties, context: Context)¶

Bases: InputExtractor

Uses TabularInputData that has previously been loaded via an input reader.

PROPERTIES = TabularProperties(name='Marginal probability calibration model', file_name='marginal_probability_calibration_model', has_header=True)¶

class mlrl.boosting.testbed.experiments.output.probability_calibration.writer.ProbabilityCalibrationModelWriter(*extractors: DataExtractor, input_data: InputData | None = None)¶

Bases: ResultWriter, ABC

Allows writing textual representations of probability calibration models to one or several sinks.

class DefaultExtractor¶

Bases: DataExtractor, ABC

An abstract base class for all classes that extract probability calibration models that are stored as part of a rule model.

extract_data(state: ExperimentState, _: list[Sink]) → list[tuple[ExperimentState, OutputData]]¶: See mlrl.testbed.experiments.output.writer.DataExtractor.extract_data()

class InputExtractor(properties: TabularProperties, context: Context)¶

Bases: TabularDataExtractor, ABC

An abstract base class for all classes that use TabularInputData that has previously been loaded via an input reader.

extract_data(state: ExperimentState, sinks: list[Sink]) → list[tuple[ExperimentState, OutputData]]¶: See mlrl.testbed.experiments.output.writer.DataExtractor.extract_data()