mlrl.testbed.experiments.output.evaluation.evaluation_result module

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

Provides classes for representing evaluation results that are part of output data.

class mlrl.testbed.experiments.output.evaluation.evaluation_result.AggregatedEvaluationResult(evaluation_by_dataset: dict[str, Table])

Bases: TabularOutputData

Stores evaluation results that have been aggregated across several experiments.

AGGREGATION_MEASURES = [<mlrl.testbed.experiments.output.evaluation.measures.AggregationMeasure object>]

COLUMN_DATASET = 'Dataset'

COLUMN_PREFIX_PARAMETER = 'Parameter'

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

OPTION_ENABLE_ALL = 'enable_all'

OPTION_RANK = 'rank'

PROPERTIES = TabularProperties(name='Evaluation results of several experiments', file_name='aggregated_evaluation', has_header=True)

to_table(options: Options, **kwargs) → Table | None

See mlrl.testbed.experiments.output.data.TabularOutputData.to_table()

to_text(options: Options, **kwargs) → str | None

See mlrl.testbed.experiments.output.data.TextualOutputData.to_text()

class mlrl.testbed.experiments.output.evaluation.evaluation_result.EvaluationResult

Bases: TabularOutputData, ABC

An abstract base class for all output data that stores evaluation results.

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

PROPERTIES = TabularProperties(name='Evaluation result', file_name='evaluation', has_header=True)