spaCy/spacy/en/__init__.py

35 lines
976 B
Python
Raw Normal View History

2017-03-12 15:07:28 +03:00
# coding: utf8
from __future__ import unicode_literals
2015-07-07 15:00:07 +03:00
2015-08-25 16:37:30 +03:00
from ..language import Language
from ..lemmatizer import Lemmatizer
from ..vocab import Vocab
from ..tokenizer import Tokenizer
from ..attrs import LANG
from ..deprecated import fix_glove_vectors_loading
2016-12-18 18:54:19 +03:00
from .language_data import *
2015-08-25 16:37:30 +03:00
class English(Language):
2015-12-28 18:54:03 +03:00
lang = 'en'
2016-09-24 21:26:17 +03:00
class Defaults(Language.Defaults):
lex_attr_getters = dict(Language.Defaults.lex_attr_getters)
lex_attr_getters[LANG] = lambda text: 'en'
2016-09-24 21:26:17 +03:00
tokenizer_exceptions = TOKENIZER_EXCEPTIONS
tag_map = TAG_MAP
stop_words = STOP_WORDS
2017-03-15 17:23:22 +03:00
morph_rules = dict(MORPH_RULES)
lemma_rules = dict(LEMMA_RULES)
lemma_index = dict(LEMMA_INDEX)
lemma_exc = dict(LEMMA_EXC)
def __init__(self, **overrides):
# Special-case hack for loading the GloVe vectors, to support <1.0
overrides = fix_glove_vectors_loading(overrides)
Language.__init__(self, **overrides)