def get_config():
"""
Get key configuration info about dev environment: OS, python, spacy, and textacy.
Returns:
dict
"""
from spacy.about import __version__ as spacy_version
from spacy.util import get_data_path
from .about import __version__ as textacy_version
return {
"platform":
sys.platform,
"python":
sys.version,
"spacy":
spacy_version,
"spacy_models": [
d.parts[-1] for d in get_data_path().iterdir()
if (d.is_dir() or d.is_symlink())
and d.parts[-1] not in {"__cache__", "__pycache__"}
],
"textacy":
textacy_version,
}
def model_installed(name):
'''
Check if spaCy language model is installed
:param name:
:return:
'''
data_path = util.get_data_path()
model_path = data_path / name
return model_path.exists()
def model_installed(name):
'''
Check if spaCy language model is installed
:param name:
:return:
'''
data_path = util.get_data_path()
model_path = data_path / name
return model_path.exists()
def model_installed(name):
'''
Check if spaCy language model is installed
:param name:
:return:
'''
data_path = util.get_data_path()
model_name = resolve_model_name(name)
model_path = data_path / model_name
if not model_path.exists():
lang_name = util.get_lang_class(name).lang
return False
return True
def vocab():
path = os.environ.get('SPACY_DATA')
if path is None:
path = util.match_best_version('en', None, util.get_data_path())
else:
path = util.match_best_version('en', None, path)
vocab = English.Defaults('en', path).Vocab()
lex = vocab['dog']
assert vocab[vocab.strings['dog']].orth_ == 'dog'
lex = vocab['the']
lex = vocab['quick']
lex = vocab['jumped']
return vocab
def vocab():
path = os.environ.get('SPACY_DATA')
if path is None:
path = util.match_best_version('en', None, util.get_data_path())
else:
path = util.match_best_version('en', None, path)
vocab = English.Defaults.create_vocab()
lex = vocab['dog']
assert vocab[vocab.strings['dog']].orth_ == 'dog'
lex = vocab['the']
lex = vocab['quick']
lex = vocab['jumped']
return vocab
def get_config():
"""Helper function to get relevant config info, especially when debugging."""
from spacy.about import __version__ as spacy_version
from spacy.util import get_data_path
from textacy import __version__ as textacy_version
return {
'python': sys.version,
'platform': sys.platform,
'textacy': textacy_version,
'spacy': spacy_version,
'spacy_models': [
d.parts[-1] for d in get_data_path().iterdir()
if (d.is_dir() or d.is_symlink()) and
d.parts[-1] not in {'__cache__', '__pycache__'}]
}
def model_installed(name):
"""Check if spaCy language model is installed.
From https://github.com/explosion/spaCy/blob/master/spacy/util.py
:param name:
:return:
"""
data_path = util.get_data_path()
if not data_path or not data_path.exists():
raise IOError(f"Can't find spaCy data path: {data_path}")
if name in {d.name for d in data_path.iterdir()}:
return True
if Spacy.is_package(name): # installed as package
return True
if Path(name).exists(): # path to model data directory
return True
return False
def model_installed(name):
'''
Check if spaCy language model is installed
From https://github.com/explosion/spaCy/blob/master/spacy/util.py
:param name:
:return:
'''
data_path = util.get_data_path()
if not data_path or not data_path.exists():
raise IOError("Can't find spaCy data path: %s" % str(data_path))
if name in set([d.name for d in data_path.iterdir()]):
return True
if SpacyTokenizer.is_package(name): # installed as package
return True
if Path(name).exists(): # path to model data directory
return True
return False
def model_installed(name):
'''
Check if spaCy language model is installed
From https://github.com/explosion/spaCy/blob/master/spacy/util.py
:param name:
:return:
'''
data_path = util.get_data_path()
if not data_path or not data_path.exists():
raise IOError("Can't find spaCy data path: %s" % str(data_path))
if name in set([d.name for d in data_path.iterdir()]):
return True
if SpacyTokenizer.is_package(name): # installed as package
return True
if Path(name).exists(): # path to model data directory
return True
return False
def model_installed(name: str):
"""Check if spaCy language model is installed
From https://github.com/explosion/spaCy/blob/master/spacy/util.py
Arguments:
name {str} -- Name of package
Returns:
[bool] -- True if installed package, False if not.
"""
data_path = util.get_data_path()
if not data_path or not data_path.exists():
raise IOError("Can't find spaCy data path: %s" % str(data_path))
if name in set([d.name for d in data_path.iterdir()]):
return True
if SpacyAnnotator.is_package(name): # installed as package
return True
if Path(name).exists(): # path to model data directory
return True
return False
def path():
return util.match_best_version(
'en', None, os.environ.get('SPACY_DATA', util.get_data_path()))
def path():
if 'SPACY_DATA' in os.environ:
return pathlib.Path(os.environ['SPACY_DATA'])
else:
return util.match_best_version('en', None, util.get_data_path())
def path():
if 'SPACY_DATA' in os.environ:
return pathlib.Path(os.environ['SPACY_DATA'])
else:
return util.match_best_version('en', None, util.get_data_path())
from spacy.util import get_data_path data_path = get_data_path() print([p for p in data_path.iterdir()]) print([p for p in data_path.iterdir() if p.is_dir()]) print([p.parts[-1] for p in data_path.iterdir()]) print([p.parts[-1] for p in data_path.iterdir() if p.is_dir()])