spaCy/spacy/lemmatizer.py

92 lines
3.0 KiB
Python
Raw Normal View History

from __future__ import unicode_literals, print_function
2015-01-04 22:01:32 +03:00
import codecs
2016-09-24 21:26:17 +03:00
import pathlib
2014-12-23 07:16:57 +03:00
2016-10-20 22:23:26 +03:00
import ujson as json
from .en.lemmatizer import INDEX, EXC, RULES
from .symbols import POS, NOUN, VERB, ADJ, PUNCT
from .symbols import VerbForm_inf, VerbForm_none
2014-12-23 07:16:57 +03:00
class Lemmatizer(object):
@classmethod
2016-12-18 17:50:09 +03:00
def load(cls, path, rules=None):
index = dict(INDEX)
exc = dict(EXC)
rules = dict(RULES)
return cls(index, exc, rules)
2014-12-23 07:16:57 +03:00
def __init__(self, index, exceptions, rules):
self.index = index
self.exc = exceptions
self.rules = rules
2015-08-25 16:46:19 +03:00
def __call__(self, string, univ_pos, morphology=None):
2016-09-27 15:16:22 +03:00
if univ_pos == NOUN:
univ_pos = 'noun'
elif univ_pos == VERB:
univ_pos = 'verb'
elif univ_pos == ADJ:
univ_pos = 'adj'
elif univ_pos == PUNCT:
univ_pos = 'punct'
# See Issue #435 for example of where this logic is requied.
if self.is_base_form(univ_pos, morphology):
return set([string.lower()])
2016-09-27 18:47:05 +03:00
lemmas = lemmatize(string, self.index.get(univ_pos, {}),
self.exc.get(univ_pos, {}),
self.rules.get(univ_pos, []))
2015-09-08 16:38:03 +03:00
return lemmas
2014-12-23 07:16:57 +03:00
def is_base_form(self, univ_pos, morphology=None):
'''Check whether we're dealing with an uninflected paradigm, so we can
avoid lemmatization entirely.'''
morphology = {} if morphology is None else morphology
others = [key for key in morphology if key not in (POS, 'number', 'pos', 'verbform')]
true_morph_key = morphology.get('morph', 0)
2016-09-27 18:47:05 +03:00
if univ_pos == 'noun' and morphology.get('number') == 'sing' and not others:
return True
2016-09-27 18:47:05 +03:00
elif univ_pos == 'verb' and morphology.get('verbform') == 'inf' and not others:
return True
elif true_morph_key in (VerbForm_inf, VerbForm_none):
return True
else:
return False
2014-12-23 07:16:57 +03:00
def noun(self, string, morphology=None):
return self(string, 'noun', morphology)
2014-12-23 07:16:57 +03:00
def verb(self, string, morphology=None):
return self(string, 'verb', morphology)
2014-12-23 07:16:57 +03:00
def adj(self, string, morphology=None):
return self(string, 'adj', morphology)
def punct(self, string, morphology=None):
return self(string, 'punct', morphology)
2014-12-23 07:16:57 +03:00
def lemmatize(string, index, exceptions, rules):
string = string.lower()
forms = []
# TODO: Is this correct? See discussion in Issue #435.
#if string in index:
# forms.append(string)
2014-12-23 07:16:57 +03:00
forms.extend(exceptions.get(string, []))
2017-03-01 23:44:17 +03:00
oov_forms = []
2014-12-23 07:16:57 +03:00
for old, new in rules:
if string.endswith(old):
form = string[:len(string) - len(old)] + new
if not form:
pass
elif form in index or not form.isalpha():
2014-12-23 07:16:57 +03:00
forms.append(form)
2017-03-01 23:44:17 +03:00
else:
oov_forms.append(form)
2014-12-23 07:16:57 +03:00
if not forms:
2017-03-01 23:44:17 +03:00
forms.extend(oov_forms)
if not forms:
forms.append(string)
2014-12-23 07:16:57 +03:00
return set(forms)