explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2024-11-11 04:08:09 +03:00
Code Issues Packages Projects Releases Wiki Activity

Merge pull request #726 from magnusburton/master

Browse Source 
Added Swedish abbreviations as token exceptions
This commit is contained in:
Ines Montani 2017-01-09 14:58:15 +01:00 committed by GitHub
commit 363f09e68c
1 changed files with 63 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

64
spacy/sv/tokenizer_exceptions.py
View File

@ -6,7 +6,69 @@ from ..language_data import PRON_LEMMA
TOKENIZER_EXCEPTIONS = {
"jan.": [
{ORTH: "jan.", LEMMA: "januari"}
],
"febr.": [
{ORTH: "febr.", LEMMA: "februari"}
],
"feb.": [
{ORTH: "feb.", LEMMA: "februari"}
],
"apr.": [
{ORTH: "apr.", LEMMA: "april"}
],
"jun.": [
{ORTH: "jun.", LEMMA: "juni"}
],
"jul.": [
{ORTH: "jul.", LEMMA: "juli"}
],
"aug.": [
{ORTH: "aug.", LEMMA: "augusti"}
],
"sept.": [
{ORTH: "sept.", LEMMA: "september"}
],
"sep.": [
{ORTH: "sep.", LEMMA: "september"}
],
"okt.": [
{ORTH: "okt.", LEMMA: "oktober"}
],
"nov.": [
{ORTH: "nov.", LEMMA: "november"}
],
"dec.": [
{ORTH: "dec.", LEMMA: "december"}
],
"mån.": [
{ORTH: "mån.", LEMMA: "måndag"}
],
"tis.": [
{ORTH: "tis.", LEMMA: "tisdag"}
],
"ons.": [
{ORTH: "ons.", LEMMA: "onsdag"}
],
"tors.": [
{ORTH: "tors.", LEMMA: "torsdag"}
],
"fre.": [
{ORTH: "fre.", LEMMA: "fredag"}
],
"lör.": [
{ORTH: "lör.", LEMMA: "lördag"}
],
"sön.": [
{ORTH: "sön.", LEMMA: "söndag"}
],
"sthlm": [
{ORTH: "sthlm", LEMMA: "Stockholm"}
],
"gbg": [
{ORTH: "gbg", LEMMA: "Göteborg"}
]
}