Rooba
/
spaCy
mirror of https://github.com/explosion/spaCy.git
1
1
Fork 0
Code Issues Packages Projects Releases Wiki Activity

Remove 'fix_pretrained_vectors_name' hack

This commit is contained in:
Matthew Honnibal 2020-08-25 14:37:45 +02:00
parent ef43152af4
commit 4f82a02b70
1 changed files with 1 additions and 25 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

26
spacy/language.py
View File

@ -1538,7 +1538,6 @@ class Language:
def deserialize_vocab(path: Path) -> None: def deserialize_vocab(path: Path) -> None:
if path.exists(): if path.exists():
self.vocab.from_disk(path) self.vocab.from_disk(path)
_fix_pretrained_vectors_name(self)
path = util.ensure_path(path) path = util.ensure_path(path)
deserializers = {} deserializers = {}
@ -1605,14 +1604,10 @@ class Language:
# from self.vocab.vectors, so set the name directly # from self.vocab.vectors, so set the name directly
self.vocab.vectors.name = data.get("vectors", {}).get("name") self.vocab.vectors.name = data.get("vectors", {}).get("name")
def deserialize_vocab(b):
self.vocab.from_bytes(b)
_fix_pretrained_vectors_name(self)
deserializers = {} deserializers = {}
deserializers["config.cfg"] = lambda b: self.config.from_bytes(b) deserializers["config.cfg"] = lambda b: self.config.from_bytes(b)
deserializers["meta.json"] = deserialize_meta deserializers["meta.json"] = deserialize_meta
deserializers["vocab"] = deserialize_vocab deserializers["vocab"] = self.vocab.from_bytes
deserializers["tokenizer"] = lambda b: self.tokenizer.from_bytes( deserializers["tokenizer"] = lambda b: self.tokenizer.from_bytes(
b, exclude=["vocab"] b, exclude=["vocab"]
) )
@ -1646,25 +1641,6 @@ class FactoryMeta:
default_score_weights: Optional[Dict[str, float]] = None # noqa: E704 default_score_weights: Optional[Dict[str, float]] = None # noqa: E704
def _fix_pretrained_vectors_name(nlp: Language) -> None:
# TODO: Replace this once we handle vectors consistently as static
# data
if "vectors" in nlp.meta and "name" in nlp.meta["vectors"]:
nlp.vocab.vectors.name = nlp.meta["vectors"]["name"]
elif not nlp.vocab.vectors.size:
nlp.vocab.vectors.name = None
elif "name" in nlp.meta and "lang" in nlp.meta:
vectors_name = f"{nlp.meta['lang']}_{nlp.meta['name']}.vectors"
nlp.vocab.vectors.name = vectors_name
else:
raise ValueError(Errors.E092)
for name, proc in nlp.pipeline:
if not hasattr(proc, "cfg") or not isinstance(proc.cfg, dict):
continue
proc.cfg.setdefault("deprecation_fixes", {})
proc.cfg["deprecation_fixes"]["vectors_name"] = nlp.vocab.vectors.name
class DisabledPipes(list): class DisabledPipes(list):
"""Manager for temporary pipeline disabling.""" """Manager for temporary pipeline disabling."""