mirror of https://github.com/explosion/spaCy.git
45 lines
1.0 KiB
Python
45 lines
1.0 KiB
Python
import pathlib
|
|
|
|
from .util import set_lang_class, get_lang_class
|
|
|
|
from . import en
|
|
from . import de
|
|
from . import zh
|
|
|
|
|
|
try:
|
|
basestring
|
|
except NameError:
|
|
basestring = str
|
|
|
|
|
|
|
|
set_lang_class(en.English.lang, en.English)
|
|
set_lang_class(de.German.lang, de.German)
|
|
set_lang_class(zh.Chinese.lang, zh.Chinese)
|
|
|
|
|
|
def load(name, vocab=True, tokenizer=True, parser=True, tagger=True, entity=True,
|
|
matcher=True, serializer=True, vectors=True, pipeline=True, via=None):
|
|
if via is None:
|
|
via = util.get_data_path()
|
|
|
|
target_name, target_version = util.split_data_name(name)
|
|
path = util.match_best_version(target_name, target_version, via)
|
|
|
|
if isinstance(vectors, basestring):
|
|
vectors = util.match_best_version(vectors, None, via)
|
|
|
|
cls = get_lang_class(target_name)
|
|
return cls(
|
|
path,
|
|
vectors=vectors,
|
|
vocab=vocab,
|
|
tokenizer=tokenizer,
|
|
tagger=tagger,
|
|
parser=parser,
|
|
entity=entity,
|
|
matcher=matcher,
|
|
pipeline=pipeline,
|
|
serializer=serializer)
|