explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-01-12 18:26:30 +03:00
Code Issues Packages Projects Releases Wiki Activity

Modifications/updates to Portuguese tokenization (#5203)

Browse Source 
Modifications to Portuguese tokenization for UD_Portuguese-Bosque.
Instead of splitting contactions as exceptions, they are kept as merged
tokens.
This commit is contained in:
adrianeboyd 2020-03-25 11:27:53 +01:00 committed by GitHub
parent 4117a5c705
commit 923a453449
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 19 additions and 41 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

60
spacy/lang/pt/tokenizer_exceptions.py
View File

@ -4,69 +4,47 @@ from __future__ import unicode_literals
from ...symbols import ORTH, NORM from ...symbols import ORTH, NORM
_exc = { _exc = {}
"às": [{ORTH: "à", NORM: "a"}, {ORTH: "s", NORM: "as"}],
"ao": [{ORTH: "a"}, {ORTH: "o"}],
"aos": [{ORTH: "a"}, {ORTH: "os"}],
"àquele": [{ORTH: "à", NORM: "a"}, {ORTH: "quele", NORM: "aquele"}],
"àquela": [{ORTH: "à", NORM: "a"}, {ORTH: "quela", NORM: "aquela"}],
"àqueles": [{ORTH: "à", NORM: "a"}, {ORTH: "queles", NORM: "aqueles"}],
"àquelas": [{ORTH: "à", NORM: "a"}, {ORTH: "quelas", NORM: "aquelas"}],
"àquilo": [{ORTH: "à", NORM: "a"}, {ORTH: "quilo", NORM: "aquilo"}],
"aonde": [{ORTH: "a"}, {ORTH: "onde"}],
}
# Contractions
_per_pron = ["ele", "ela", "eles", "elas"]
_dem_pron = [
"este",
"esta",
"estes",
"estas",
"isto",
"esse",
"essa",
"esses",
"essas",
"isso",
"aquele",
"aquela",
"aqueles",
"aquelas",
"aquilo",
]
_und_pron = ["outro", "outra", "outros", "outras"]
_adv = ["aqui", "aí", "ali", "além"]
for orth in _per_pron + _dem_pron + _und_pron + _adv:
_exc["d" + orth] = [{ORTH: "d", NORM: "de"}, {ORTH: orth}]
for orth in _per_pron + _dem_pron + _und_pron:
_exc["n" + orth] = [{ORTH: "n", NORM: "em"}, {ORTH: orth}]
for orth in [ for orth in [
"Adm.", "Adm.",
"Art.",
"art.",
"Av.",
"av.",
"Cia.",
"dom.",
"Dr.", "Dr.",
"dr.",
"e.g.", "e.g.",
"E.g.", "E.g.",
"E.G.", "E.G.",
"e/ou",
"ed.",
"eng.",
"etc.",
"Fund.",
"Gen.", "Gen.",
"Gov.", "Gov.",
"i.e.", "i.e.",
"I.e.", "I.e.",
"I.E.", "I.E.",
"Inc.",
"Jr.", "Jr.",
"km/h",
"Ltd.", "Ltd.",
"Mr.",
"p.m.", "p.m.",
"Ph.D.", "Ph.D.",
"Rep.", "Rep.",
"Rev.", "Rev.",
"S/A",
"Sen.", "Sen.",
"Sr.", "Sr.",
"sr.",
"Sra.", "Sra.",
"sra.",
"vs.", "vs.",
"tel.", "tel.",
"pág.", "pág.",