explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2025-07-04 20:03:13 +03:00
Code Issues Packages Projects Releases Wiki Activity

Fix overwriting of lexical attributes when loading vectors during training

Browse Source
This commit is contained in:
Matthew Honnibal 2018-02-17 18:11:11 +01:00
parent c0caf7cf27
commit 262d0a3148
1 changed files with 10 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
spacy/cli/train.py
View File

@ -7,6 +7,7 @@ import tqdm
from thinc.neural._classes.model import Model from thinc.neural._classes.model import Model
from timeit import default_timer as timer from timeit import default_timer as timer
from ..attrs import PROB, IS_OOV, CLUSTER, LANG
from ..gold import GoldCorpus, minibatch from ..gold import GoldCorpus, minibatch
from ..util import prints from ..util import prints
from .. import util from .. import util
@ -90,6 +91,15 @@ def train(lang, output_dir, train_data, dev_data, n_iter=30, n_sents=0,
nlp.meta.update(meta) nlp.meta.update(meta)
if vectors: if vectors:
util.load_model(vectors, vocab=nlp.vocab) util.load_model(vectors, vocab=nlp.vocab)
for lex in nlp.vocab:
values = {}
for attr, func in nlp.vocab.lex_attr_getters.items():
# These attrs are expected to be set by data. Others should
# be set by calling the language functions.
if attr not in (CLUSTER, PROB, IS_OOV, LANG):
values[lex.vocab.strings[attr]] = func(lex.orth_)
lex.set_attrs(**values)
lex.is_oov = False
for name in pipeline: for name in pipeline:
nlp.add_pipe(nlp.create_pipe(name), name=name) nlp.add_pipe(nlp.create_pipe(name), name=name)
optimizer = nlp.begin_training(lambda: corpus.train_tuples, device=use_gpu) optimizer = nlp.begin_training(lambda: corpus.train_tuples, device=use_gpu)