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

* Refactor init_model to accomodate other languages

This commit is contained in:
Matthew Honnibal 2015-08-26 19:14:05 +02:00
parent 494da25872
commit dc13edd7cb
1 changed files with 3 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

6
bin/init_model.py
View File

@ -151,7 +151,7 @@ def setup_vocab(get_lex_attr, src_dir, dst_dir):
write_binary_vectors(str(vectors_src), str(dst_dir / 'vec.bin')) write_binary_vectors(str(vectors_src), str(dst_dir / 'vec.bin'))
else: else:
print("Warning: Word vectors file not found") print("Warning: Word vectors file not found")
vocab = Vocab(data_dir=None, get_lex_attr=get_lex_attr) vocab = Vocab(get_lex_attr=get_lex_attr)
clusters = _read_clusters(src_dir / 'clusters.txt') clusters = _read_clusters(src_dir / 'clusters.txt')
probs, oov_prob = _read_probs(src_dir / 'words.sgt.prob') probs, oov_prob = _read_probs(src_dir / 'words.sgt.prob')
if not probs: if not probs:
@ -183,8 +183,8 @@ def setup_vocab(get_lex_attr, src_dir, dst_dir):
def main(lang_id, lang_data_dir, corpora_dir, model_dir): def main(lang_id, lang_data_dir, corpora_dir, model_dir):
languages = { languages = {
'en': spacy.en.get_lex_attr, 'en': spacy.en.English.default_lex_attrs(),
'de': spacy.en.get_lex_attr 'de': spacy.de.Deutsch.default_lex_attrs()
} }
model_dir = Path(model_dir) model_dir = Path(model_dir)