Rooba
/
spaCy
mirror of https://github.com/explosion/spaCy.git
1
1
Fork 0
Code Issues Packages Projects Releases Wiki Activity

add dot after some abbrevs

This commit is contained in:
thjbbvlt 2024-03-26 11:25:43 +01:00
parent b0b66fbab4
commit 54fcc5bba2
1 changed files with 4 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
spacy/lang/fr/tokenizer_exceptions.py
View File

@ -4,15 +4,19 @@ from ..tokenizer_exceptions import BASE_EXCEPTIONS
_exc = {
"St": [{ORTH: "St", NORM: "Saint"}],
"St.": [{ORTH: "St.", NORM: "Saint"}],
"Ste": [{ORTH: "Ste", NORM: "Sainte"}],
"Mme": [{ORTH: "Mme", NORM: "Madame"}],
"Mr": [{ORTH: "Mr", NORM: "Monsieur"}],
"Mr.": [{ORTH: "Mr.", NORM: "Monsieur"}],
"M.": [{ORTH: "M.", NORM: "Monsieur"}],
"Mlle": [{ORTH: "Mlle", NORM: "Mademoiselle"}],
"Dr": [{ORTH: "Dr", NORM: "Docteur"}],
"Dr.": [{ORTH: "Dr.", NORM: "Docteur"}],
"Dresse": [{ORTH: "Dresse", NORM: "Doctoresse"}],
"Drsse": [{ORTH: "Drsse", NORM: "Doctoresse"}],
"etc": [{ORTH: "etc", NORM: "etcaetera"}],
"etc.": [{ORTH: "etc.", NORM: "etcaetera"}],
# months
"jan.": [{ORTH: "jan.", NORM: "janvier"}],
"janv.": [{ORTH: "janv.", NORM: "janvier"}],