.datasets#

The datasets module makes it easy to download and import several corpus build for temporal information extraction.

Currently, the supported datasets are the following:

  • AQUAINT

  • EventTime

  • GraphEve

  • MATRES

  • MeanTime_English

  • MeanTime_Spanish

  • MeanTime_Dutch

  • MeanTime_Italian

  • Platinum

  • TCR

  • TDDiscourse

  • TempEval_2_Chinese

  • TempEval_2_English

  • TempEval_2_French

  • TempEval_2_Italian

  • TempEval_2_Korean

  • TempEval_2_Spanish

  • TempEval_3

  • TimeBank_1.2

  • TimeBank_Dense

  • TimeBankPT

  • TimeBank

  • TrainT

tieval.datasets.download(dataset: str, path: Path) → None#

Download corpus.

tieval.datasets.read(dataset: str, path: Path | None = None) → Dataset#

Load temporally annotated dataset.

.readers.dataset#

class tieval.datasets.readers.dataset.XMLDatasetReader(doc_reader: AncientTimeDocumentReader | TempEval3DocumentReader | TimeBank12DocumentReader | MeanTimeDocumentReader | GraphEveDocumentReader | TempEval2DocumentReader | TCRDocumentReader | TempEval2FrenchDocumentReader | TimeBankPTDocumentReader | KRAUTSDocumentReader | NarrativeContainerDocumentReader | WikiWarsDocumentReader | FRTimeBankDocumentReader | ProfessorHeidelTimeDocumentReader)#

Bases: object

Handles the process of reading any temporally annotated dataset stored with .tml or .xml extension.

.readers.document#