List of formats

Documents, annotations and exports can be downloaded/created in different formats.

INCEpTION doesn't specify in their documentation which formats are supported, but the following have been found and included in pycaprio:

  • bin: UIMA binary CAS
  • bioc: BioC XML
  • conll2000: CoNLL 2000
  • conll2002: CoNLL 2002
  • conll2003: CoNLL 2003
  • conll2006: CoNLL 2006
  • conll2009: CoNLL 2009
  • conll2012: CoNLL 2012
  • conllcorenlp: CoNLL CoreNLP
  • conllu: CoNLL-U
  • ctsv: WebAnno TSV v2
  • ctsv3: WebAnno TSV v3.3
  • dkpro-core-tei: TEI P5 XML (legacy)
  • dkpro-core-uima-inline-xml: Inline XML
  • dkpro-core-xml-document: XML (generic)
  • docx: DOCX
  • html: HTML (legacy)
  • htmldoc: HTML
  • htmldoc-zip: HTML (ZIP)
  • imscwb: Corpus Workbench Format (aka VRT)
  • json: UIMA CAS JSON (legacy)
  • jsoncas: UIMA CAS JSON 0.4.0
  • jsoncas-struct: UIMA CAS JSON 0.4.0 (XML/PDF structure)
  • lif: LAPPS Interchange Format (LIF)
  • mhtml: MHTML (Web archive)
  • nif: NLP Interchange Format (NIF)
  • pdf2: PDF
  • pdf2JsonCas: PDF (with embedded JSON CAS)
  • pdf2XmiCas: PDF (with embedded XMI CAS)
  • perseus_2.1: Perseus treebank XML 2.1
  • pubannotation-annotations: PubAnnotation Document with Annotations (JSON)
  • pubannotation-sections: PubAnnotation Document with Sections (JSON)
  • rdfcas: UIMA CAS RDF
  • tcf: WebLicht TCF
  • text: Plain text (DEFAULT)
  • textlines: Plain text (one sentence per line)
  • pretokenized-textlines: Plain text (space-separated tokens, one sentence per line)
  • tsv: WebAnno TSV v1
  • xmi: UIMA CAS XMI (XML 1.0)
  • xmi-struct: UIMA CAS XMI (XML 1.0 + XML/PDF structure)
  • xmi-xml1.1: UIMA CAS XMI (XML 1.1)

You can find a class with all the formats in pycaprio.core.mappings.InceptionFormat:

from pycaprio.core.mappings import InceptionFormat

InceptionFormat.DEFAULT # Defaults to `text`
InceptionFormat.TEI
...