nessie.detectors.retag

Module Contents

Classes

Retag

Halteren, Hans van. “The Detection of Innessie in Manually Tagged Text.”

class nessie.detectors.retag.Retag

Bases: nessie.detectors.error_detector.ModelBasedDetector

Halteren, Hans van. “The Detection of Innessie in Manually Tagged Text.” In Proceedings of the COLING-2000 Workshop on Linguistically Interpreted Corpora, 48–55. Centre Universitaire, Luxembourg: International Committee on Computational Linguistics, 2000. https://www.aclweb.org/anthology/W00-1907.

correct(self, predictions: numpy.ndarray, **kwargs) → numpy.ndarray

error_detector_kind(self)

score(self, labels: nessie.types.StringArray, predictions: nessie.types.StringArray, **kwargs) → numpy.typing.NDArray[bool]

Flags the input if the noisy labels disagree with the predictions of a ML model.

Parameters

• labels – a (num_instances, ) string sequence containing the noisy label for each instance

• predictions – a (num_instances, ) numpy array obtained from a machine learning model

Returns

a (num_instances,) numpy array of bools containing the flags after using retag

supports_correction(self) → bool