explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-01-12 02:06:31 +03:00
Code Issues Packages Projects Releases Wiki Activity

Add a few more Danish tokenizer exceptions

Browse Source
This commit is contained in:
Adriane Boyd 2020-02-26 14:59:03 +01:00
parent 1c212215cd
commit 9f740a9891
1 changed files with 12 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

12
spacy/lang/da/tokenizer_exceptions.py
View File

@ -70,6 +70,7 @@ for orth in [
"A/S", "A/S",
"B.C.", "B.C.",
"BK.", "BK.",
"B.T.",
"Dr.", "Dr.",
"Boul.", "Boul.",
"Chr.", "Chr.",
@ -79,6 +80,7 @@ for orth in [
"Hf.", "Hf.",
"i/s", "i/s",
"I/S", "I/S",
"Inc.",
"Kprs.", "Kprs.",
"L.A.", "L.A.",
"Ll.", "Ll.",
@ -149,6 +151,7 @@ for orth in [
"bygn.", "bygn.",
"c/o", "c/o",
"ca.", "ca.",
"cm.",
"cand.", "cand.",
"d.d.", "d.d.",
"d.m.", "d.m.",
@ -172,10 +175,12 @@ for orth in [
"dl.", "dl.",
"do.", "do.",
"dobb.", "dobb.",
"dr.",
"dr.h.c", "dr.h.c",
"dr.phil.", "dr.phil.",
"ds.", "ds.",
"dvs.", "dvs.",
"d.v.s.",
"e.b.", "e.b.",
"e.l.", "e.l.",
"e.o.", "e.o.",
@ -297,10 +302,14 @@ for orth in [
"kap.", "kap.",
"kbh.", "kbh.",
"kem.", "kem.",
"kg.",
"kgs.",
"kgl.", "kgl.",
"kl.", "kl.",
"kld.", "kld.",
"km.",
"km/t", "km/t",
"km/t.",
"knsp.", "knsp.",
"komm.", "komm.",
"kons.", "kons.",
@ -311,6 +320,7 @@ for orth in [
"kt.", "kt.",
"ktr.", "ktr.",
"kv.", "kv.",
"kvm.",
"kvt.", "kvt.",
"l.c.", "l.c.",
"lab.", "lab.",
@ -357,6 +367,7 @@ for orth in [
"nto.", "nto.",
"nuv.", "nuv.",
"o/m", "o/m",
"o/m.",
"o.a.", "o.a.",
"o.fl.", "o.fl.",
"o.h.", "o.h.",
@ -526,6 +537,7 @@ for orth in [
"vejl.", "vejl.",
"vh.", "vh.",
"vha.", "vha.",
"vind.",
"vs.", "vs.",
"vsa.", "vsa.",
"vær.", "vær.",