mirror of
https://github.com/explosion/spaCy.git
synced 2024-09-22 11:59:14 +03:00
eae35e9b27
- add files to specify rules for German tokenization - change generate_specials.py to generate from an external file (abbrev.de.tab) - copy gazetteer.json from lang_data/en/ - init_model.py - change doc freq threshold to 0 - add train_german_tagger.py - expects conll09-formatted input
335 lines
12 KiB
Python
335 lines
12 KiB
Python
# coding=utf8
|
||
import json
|
||
import io
|
||
import itertools
|
||
|
||
contractions = {}
|
||
|
||
# contains the lemmas, parts of speech, number, and tenspect of
|
||
# potential tokens generated after splitting contractions off
|
||
token_properties = {}
|
||
|
||
# contains starting tokens with their potential contractions
|
||
# each potential contraction has a list of exceptions
|
||
# lower - don't generate the lowercase version
|
||
# upper - don't generate the uppercase version
|
||
# contrLower - don't generate the lowercase version with apostrophe (') removed
|
||
# contrUpper - dont' generate the uppercase version with apostrophe (') removed
|
||
# for example, we don't want to create the word "hell" or "Hell" from "he" + "'ll" so
|
||
# we add "contrLower" and "contrUpper" to the exceptions list
|
||
starting_tokens = {}
|
||
|
||
# other specials that don't really have contractions
|
||
# so they are hardcoded
|
||
hardcoded_specials = {
|
||
"''": [{"F": "''"}],
|
||
"\")": [{"F": "\")"}],
|
||
"\n": [{"F": "\n", "pos": "SP"}],
|
||
"\t": [{"F": "\t", "pos": "SP"}],
|
||
" ": [{"F": " ", "pos": "SP"}],
|
||
|
||
# example: Wie geht's?
|
||
"'s": [{"F": "'s", "L": "es"}],
|
||
"'S": [{"F": "'S", "L": "es"}],
|
||
|
||
# example: Haste mal 'nen Euro?
|
||
"'n": [{"F": "'n", "L": "ein"}],
|
||
"'ne": [{"F": "'ne", "L": "eine"}],
|
||
"'nen": [{"F": "'nen", "L": "einen"}],
|
||
|
||
# example: Kommen S’ nur herein!
|
||
"s'": [{"F": "s'", "L": "sie"}],
|
||
"S'": [{"F": "S'", "L": "sie"}],
|
||
|
||
# example: Da haben wir's!
|
||
"ich's": [{"F": "ich"}, {"F": "'s", "L": "es"}],
|
||
"du's": [{"F": "du"}, {"F": "'s", "L": "es"}],
|
||
"er's": [{"F": "er"}, {"F": "'s", "L": "es"}],
|
||
"sie's": [{"F": "sie"}, {"F": "'s", "L": "es"}],
|
||
"wir's": [{"F": "wir"}, {"F": "'s", "L": "es"}],
|
||
"ihr's": [{"F": "ihr"}, {"F": "'s", "L": "es"}],
|
||
|
||
# example: Die katze auf'm dach.
|
||
"auf'm": [{"F": "auf"}, {"F": "'m", "L": "dem"}],
|
||
"unter'm": [{"F": "unter"}, {"F": "'m", "L": "dem"}],
|
||
"über'm": [{"F": "über"}, {"F": "'m", "L": "dem"}],
|
||
"vor'm": [{"F": "vor"}, {"F": "'m", "L": "dem"}],
|
||
"hinter'm": [{"F": "hinter"}, {"F": "'m", "L": "dem"}],
|
||
|
||
# persons
|
||
"Fr.": [{"F": "Fr."}],
|
||
"Hr.": [{"F": "Hr."}],
|
||
"Frl.": [{"F": "Frl."}],
|
||
"Prof.": [{"F": "Prof."}],
|
||
"Dr.": [{"F": "Dr."}],
|
||
"St.": [{"F": "St."}],
|
||
"Hrgs.": [{"F": "Hrgs."}],
|
||
"Hg.": [{"F": "Hg."}],
|
||
"a.Z.": [{"F": "a.Z."}],
|
||
"a.D.": [{"F": "a.D."}],
|
||
"A.D.": [{"F": "A.D."}],
|
||
"h.c.": [{"F": "h.c."}],
|
||
"jun.": [{"F": "jun."}],
|
||
"sen.": [{"F": "sen."}],
|
||
"rer.": [{"F": "rer."}],
|
||
"Dipl.": [{"F": "Dipl."}],
|
||
"Ing.": [{"F": "Ing."}],
|
||
"Dipl.-Ing.": [{"F": "Dipl.-Ing."}],
|
||
|
||
# companies
|
||
"Co.": [{"F": "Co."}],
|
||
"co.": [{"F": "co."}],
|
||
"Cie.": [{"F": "Cie."}],
|
||
"A.G.": [{"F": "A.G."}],
|
||
"G.m.b.H.": [{"F": "G.m.b.H."}],
|
||
"i.G.": [{"F": "i.G."}],
|
||
"e.V.": [{"F": "e.V."}],
|
||
|
||
# popular german abbreviations
|
||
"ggü.": [{"F": "ggü."}],
|
||
"ggf.": [{"F": "ggf."}],
|
||
"ggfs.": [{"F": "ggfs."}],
|
||
"Gebr.": [{"F": "Gebr."}],
|
||
"geb.": [{"F": "geb."}],
|
||
"gegr.": [{"F": "gegr."}],
|
||
"erm.": [{"F": "erm."}],
|
||
"engl.": [{"F": "engl."}],
|
||
"ehem.": [{"F": "ehem."}],
|
||
"Biol.": [{"F": "Biol."}],
|
||
"biol.": [{"F": "biol."}],
|
||
"Abk.": [{"F": "Abk."}],
|
||
"Abb.": [{"F": "Abb."}],
|
||
"abzgl.": [{"F": "abzgl."}],
|
||
"Hbf.": [{"F": "Hbf."}],
|
||
"Bhf.": [{"F": "Bhf."}],
|
||
"Bf.": [{"F": "Bf."}],
|
||
"i.V.": [{"F": "i.V."}],
|
||
"inkl.": [{"F": "inkl."}],
|
||
"insb.": [{"F": "insb."}],
|
||
"z.B.": [{"F": "z.B."}],
|
||
"i.Tr.": [{"F": "i.Tr."}],
|
||
"Jhd.": [{"F": "Jhd."}],
|
||
"jur.": [{"F": "jur."}],
|
||
"lt.": [{"F": "lt."}],
|
||
"nat.": [{"F": "nat."}],
|
||
"u.a.": [{"F": "u.a."}],
|
||
"u.s.w.": [{"F": "u.s.w."}],
|
||
"Nr.": [{"F": "Nr."}],
|
||
"Univ.": [{"F": "Univ."}],
|
||
"vgl.": [{"F": "vgl."}],
|
||
"zzgl.": [{"F": "zzgl."}],
|
||
"z.Z.": [{"F": "z.Z."}],
|
||
"betr.": [{"F": "betr."}],
|
||
"ehem.": [{"F": "ehem."}],
|
||
|
||
# popular latin abbreviations
|
||
"vs.": [{"F": "vs."}],
|
||
"adv.": [{"F": "adv."}],
|
||
"Chr.": [{"F": "Chr."}],
|
||
"A.C.": [{"F": "A.C."}],
|
||
"A.D.": [{"F": "A.D."}],
|
||
"e.g.": [{"F": "e.g."}],
|
||
"i.e.": [{"F": "i.e."}],
|
||
"al.": [{"F": "al."}],
|
||
"p.a.": [{"F": "p.a."}],
|
||
"P.S.": [{"F": "P.S."}],
|
||
"q.e.d.": [{"F": "q.e.d."}],
|
||
"R.I.P.": [{"F": "R.I.P."}],
|
||
"etc.": [{"F": "etc."}],
|
||
"incl.": [{"F": "incl."}],
|
||
|
||
# popular english abbreviations
|
||
"D.C.": [{"F": "D.C."}],
|
||
"N.Y.": [{"F": "N.Y."}],
|
||
"N.Y.C.": [{"F": "N.Y.C."}],
|
||
|
||
# dates
|
||
"Jan.": [{"F": "Jan."}],
|
||
"Feb.": [{"F": "Feb."}],
|
||
"Mrz.": [{"F": "Mrz."}],
|
||
"Mär.": [{"F": "Mär."}],
|
||
"Apr.": [{"F": "Apr."}],
|
||
"Jun.": [{"F": "Jun."}],
|
||
"Jul.": [{"F": "Jul."}],
|
||
"Aug.": [{"F": "Aug."}],
|
||
"Sep.": [{"F": "Sep."}],
|
||
"Sept.": [{"F": "Sept."}],
|
||
"Okt.": [{"F": "Okt."}],
|
||
"Nov.": [{"F": "Nov."}],
|
||
"Dez.": [{"F": "Dez."}],
|
||
"Mo.": [{"F": "Mo."}],
|
||
"Di.": [{"F": "Di."}],
|
||
"Mi.": [{"F": "Mi."}],
|
||
"Do.": [{"F": "Do."}],
|
||
"Fr.": [{"F": "Fr."}],
|
||
"Sa.": [{"F": "Sa."}],
|
||
"So.": [{"F": "So."}],
|
||
|
||
# smileys
|
||
":)": [{"F": ":)"}],
|
||
"<3": [{"F": "<3"}],
|
||
";)": [{"F": ";)"}],
|
||
"(:": [{"F": "(:"}],
|
||
":(": [{"F": ":("}],
|
||
"-_-": [{"F": "-_-"}],
|
||
"=)": [{"F": "=)"}],
|
||
":/": [{"F": ":/"}],
|
||
":>": [{"F": ":>"}],
|
||
";-)": [{"F": ";-)"}],
|
||
":Y": [{"F": ":Y"}],
|
||
":P": [{"F": ":P"}],
|
||
":-P": [{"F": ":-P"}],
|
||
":3": [{"F": ":3"}],
|
||
"=3": [{"F": "=3"}],
|
||
"xD": [{"F": "xD"}],
|
||
"^_^": [{"F": "^_^"}],
|
||
"=]": [{"F": "=]"}],
|
||
"=D": [{"F": "=D"}],
|
||
"<333": [{"F": "<333"}],
|
||
":))": [{"F": ":))"}],
|
||
":0": [{"F": ":0"}],
|
||
"-__-": [{"F": "-__-"}],
|
||
"xDD": [{"F": "xDD"}],
|
||
"o_o": [{"F": "o_o"}],
|
||
"o_O": [{"F": "o_O"}],
|
||
"V_V": [{"F": "V_V"}],
|
||
"=[[": [{"F": "=[["}],
|
||
"<33": [{"F": "<33"}],
|
||
";p": [{"F": ";p"}],
|
||
";D": [{"F": ";D"}],
|
||
";-p": [{"F": ";-p"}],
|
||
";(": [{"F": ";("}],
|
||
":p": [{"F": ":p"}],
|
||
":]": [{"F": ":]"}],
|
||
":O": [{"F": ":O"}],
|
||
":-/": [{"F": ":-/"}],
|
||
":-)": [{"F": ":-)"}],
|
||
":(((": [{"F": ":((("}],
|
||
":((": [{"F": ":(("}],
|
||
":')": [{"F": ":')"}],
|
||
"(^_^)": [{"F": "(^_^)"}],
|
||
"(=": [{"F": "(="}],
|
||
"o.O": [{"F": "o.O"}],
|
||
|
||
"a.": [{"F": "a."}],
|
||
"b.": [{"F": "b."}],
|
||
"c.": [{"F": "c."}],
|
||
"d.": [{"F": "d."}],
|
||
"e.": [{"F": "e."}],
|
||
"f.": [{"F": "f."}],
|
||
"g.": [{"F": "g."}],
|
||
"h.": [{"F": "h."}],
|
||
"i.": [{"F": "i."}],
|
||
"j.": [{"F": "j."}],
|
||
"k.": [{"F": "k."}],
|
||
"l.": [{"F": "l."}],
|
||
"m.": [{"F": "m."}],
|
||
"n.": [{"F": "n."}],
|
||
"o.": [{"F": "o."}],
|
||
"p.": [{"F": "p."}],
|
||
"q.": [{"F": "q."}],
|
||
"r.": [{"F": "r."}],
|
||
"s.": [{"F": "s."}],
|
||
"t.": [{"F": "t."}],
|
||
"u.": [{"F": "u."}],
|
||
"v.": [{"F": "v."}],
|
||
"w.": [{"F": "w."}],
|
||
"x.": [{"F": "x."}],
|
||
"y.": [{"F": "y."}],
|
||
"z.": [{"F": "z."}],
|
||
}
|
||
|
||
def get_double_contractions(ending):
|
||
endings = []
|
||
|
||
ends_with_contraction = any([ending.endswith(contraction) for contraction in contractions])
|
||
|
||
while ends_with_contraction:
|
||
for contraction in contractions:
|
||
if ending.endswith(contraction):
|
||
endings.append(contraction)
|
||
ending = ending.rstrip(contraction)
|
||
ends_with_contraction = any([ending.endswith(contraction) for contraction in contractions])
|
||
|
||
endings.reverse() # reverse because the last ending is put in the list first
|
||
return endings
|
||
|
||
def get_token_properties(token, capitalize=False, remove_contractions=False):
|
||
props = dict(token_properties.get(token)) # ensure we copy the dict so we can add the "F" prop
|
||
if capitalize:
|
||
token = token.capitalize()
|
||
if remove_contractions:
|
||
token = token.replace("'", "")
|
||
|
||
props["F"] = token
|
||
return props
|
||
|
||
|
||
def create_entry(token, endings, capitalize=False, remove_contractions=False):
|
||
properties = []
|
||
properties.append(get_token_properties(token, capitalize=capitalize, remove_contractions=remove_contractions))
|
||
for e in endings:
|
||
properties.append(get_token_properties(e, remove_contractions=remove_contractions))
|
||
return properties
|
||
|
||
|
||
FIELDNAMES = ['F','L','pos']
|
||
def read_hardcoded(stream):
|
||
hc_specials = {}
|
||
for line in stream:
|
||
line = line.strip()
|
||
if line.startswith('#') or not line:
|
||
continue
|
||
key,_,rest = line.partition('\t')
|
||
values = []
|
||
for annotation in zip(*[ e.split('|') for e in rest.split('\t') ]):
|
||
values.append({ k:v for k,v in itertools.izip_longest(FIELDNAMES,annotation) if v })
|
||
hc_specials[key] = values
|
||
return hc_specials
|
||
|
||
|
||
def generate_specials():
|
||
|
||
specials = {}
|
||
|
||
for token in starting_tokens:
|
||
possible_endings = starting_tokens[token]
|
||
for ending in possible_endings:
|
||
|
||
endings = []
|
||
if ending.count("'") > 1:
|
||
endings.extend(get_double_contractions(ending))
|
||
else:
|
||
endings.append(ending)
|
||
|
||
exceptions = possible_endings[ending]
|
||
|
||
if "lower" not in exceptions:
|
||
special = token + ending
|
||
specials[special] = create_entry(token, endings)
|
||
|
||
if "upper" not in exceptions:
|
||
special = token.capitalize() + ending
|
||
specials[special] = create_entry(token, endings, capitalize=True)
|
||
|
||
if "contrLower" not in exceptions:
|
||
special = token + ending.replace("'", "")
|
||
specials[special] = create_entry(token, endings, remove_contractions=True)
|
||
|
||
if "contrUpper" not in exceptions:
|
||
special = token.capitalize() + ending.replace("'", "")
|
||
specials[special] = create_entry(token, endings, capitalize=True, remove_contractions=True)
|
||
|
||
# add in hardcoded specials
|
||
# changed it so it generates them from a file
|
||
with io.open('abbrev.de.tab','r',encoding='utf8') as abbrev_:
|
||
hc_specials = read_hardcoded(abbrev_)
|
||
specials = dict(specials, **hc_specials)
|
||
|
||
return specials
|
||
|
||
if __name__ == "__main__":
|
||
specials = generate_specials()
|
||
with open("specials.json", "w") as f:
|
||
json.dump(specials, f, sort_keys=True, indent=4, separators=(',', ': '))
|