diff options
| -rw-r--r-- | python/sandcrawler/grobid.py | 5 |
1 files changed, 5 insertions, 0 deletions
diff --git a/python/sandcrawler/grobid.py b/python/sandcrawler/grobid.py index f221830..791e0fe 100644 --- a/python/sandcrawler/grobid.py +++ b/python/sandcrawler/grobid.py @@ -95,9 +95,14 @@ class GrobidClient(object): self.host_url + "/api/processFulltextDocument", files={ "input": blob, + }, + data={ "consolidateHeader": consolidate_mode, "consolidateCitations": 0, # too expensive for now "includeRawCitations": 1, + "includeRawAffiliations": 1, + "teiCoordinates": ["ref", "figure", "persName", "formula", "biblStruct"], + "segmentSentences": 1, }, timeout=180.0, ) |