mirror of
https://github.com/explosion/spaCy.git
synced 2024-11-14 05:37:03 +03:00
30 lines
1014 B
Python
30 lines
1014 B
Python
# coding: utf8
|
||
from __future__ import unicode_literals
|
||
|
||
from .. import language_data as base
|
||
from ..language_data import update_exc, strings_to_exc, expand_exc
|
||
from ..symbols import ORTH, LEMMA
|
||
|
||
from .tag_map import TAG_MAP
|
||
from .word_sets import STOP_WORDS, NUM_WORDS
|
||
from .tokenizer_exceptions import TOKENIZER_EXCEPTIONS, ORTH_ONLY
|
||
from .morph_rules import MORPH_RULES
|
||
from .lemmatizer import RULES as LEMMA_RULES
|
||
from .lemmatizer import INDEX as LEMMA_INDEX
|
||
from .lemmatizer import EXC as LEMMA_EXC
|
||
|
||
|
||
TAG_MAP = dict(TAG_MAP)
|
||
STOP_WORDS = set(STOP_WORDS)
|
||
|
||
|
||
TOKENIZER_EXCEPTIONS = dict(TOKENIZER_EXCEPTIONS)
|
||
update_exc(TOKENIZER_EXCEPTIONS, strings_to_exc(ORTH_ONLY))
|
||
update_exc(TOKENIZER_EXCEPTIONS, expand_exc(TOKENIZER_EXCEPTIONS, "'", "’"))
|
||
update_exc(TOKENIZER_EXCEPTIONS, strings_to_exc(base.EMOTICONS))
|
||
update_exc(TOKENIZER_EXCEPTIONS, strings_to_exc(base.ABBREVIATIONS))
|
||
|
||
|
||
__all__ = ["TOKENIZER_EXCEPTIONS", "TAG_MAP", "STOP_WORDS", "MORPH_RULES",
|
||
"LEMMA_RULES", "LEMMA_INDEX", "LEMMA_EXC"]
|