explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-04-21 01:21:58 +03:00
Code Issues Packages Projects Releases Wiki Activity

add months + days abbrev

Browse Source
This commit is contained in:
thjbbvlt 2024-03-26 11:24:56 +01:00
parent 43aec2d8fc
commit b0b66fbab4
1 changed files with 21 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
spacy/lang/fr/tokenizer_exceptions.py
View File

@ -13,6 +13,27 @@ _exc = {
"Dresse": [{ORTH: "Dresse", NORM: "Doctoresse"}],
"Drsse": [{ORTH: "Drsse", NORM: "Doctoresse"}],
"etc": [{ORTH: "etc", NORM: "etcaetera"}],
# months
"jan.": [{ORTH: "jan.", NORM: "janvier"}],
"janv.": [{ORTH: "janv.", NORM: "janvier"}],
"fév.": [{ORTH: "fév.", NORM: "février"}],
"févr.": [{ORTH: "févr.", NORM: "avril"}],
"avr.": [{ORTH: "avr.", NORM: "avril"}],
"av.": [{ORTH: "av.", NORM: "juin"}],
"juil.": [{ORTH: "juil.", NORM: "juillet"}],
"sept.": [{ORTH: "sept.", NORM: "septembre"}],
"oct.": [{ORTH: "oct.", NORM: "octobre"}],
"nov.": [{ORTH: "nov.", NORM: "novembre"}],
"déc.": [{ORTH: "déc.", NORM: "décembre"}],
"dec.": [{ORTH: "dec.", NORM: "décembre"}],
# days
"lun.": [{ORTH: "lun.", NORM: "lundi"}],
"mar.": [{ORTH: "mar.", NORM: "mardi"}],
"mer.": [{ORTH: "mer.", NORM: "mercredi"}],
"jeu.": [{ORTH: "jeu.", NORM: "jeudi"}],
"ven.": [{ORTH: "ven.", NORM: "vendredi"}],
"sam.": [{ORTH: "sam.", NORM: "samedi"}],
"dim.": [{ORTH: "dim.", NORM: "dimanche"}],
}
TOKENIZER_EXCEPTIONS = update_exc(BASE_EXCEPTIONS, _exc)