mirror of
https://github.com/explosion/spaCy.git
synced 2024-11-14 05:37:03 +03:00
24 lines
590 B
Python
24 lines
590 B
Python
# encoding: utf8
|
|
from __future__ import unicode_literals, print_function
|
|
|
|
from os import path
|
|
|
|
from ..language import Language
|
|
from ..lemmatizer import Lemmatizer
|
|
from ..vocab import Vocab
|
|
from ..tokenizer import Tokenizer
|
|
from ..attrs import LANG
|
|
|
|
from .language_data import *
|
|
|
|
class English(Language):
|
|
lang = 'en'
|
|
|
|
class Defaults(Language.Defaults):
|
|
lex_attr_getters = dict(Language.Defaults.lex_attr_getters)
|
|
lex_attr_getters[LANG] = lambda text: 'en'
|
|
|
|
tokenizer_exceptions = TOKENIZER_EXCEPTIONS
|
|
tag_map = TAG_MAP
|
|
stop_words = STOP_WORDS
|