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

* Upd get_freqs script

This commit is contained in:
Matthew Honnibal 2015-07-25 21:13:41 +02:00
parent 2b7bd46508
commit 105305b4aa
1 changed files with 4 additions and 1 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
bin/get_freqs.py
View File

@ -40,6 +40,7 @@ def null_props(string):
def count_freqs(input_loc, output_loc): def count_freqs(input_loc, output_loc):
print output_loc
nlp = spacy.en.English(Parser=None, Tagger=None, Entity=None, load_vectors=False) nlp = spacy.en.English(Parser=None, Tagger=None, Entity=None, load_vectors=False)
nlp.vocab.lexeme_props_getter = null_props nlp.vocab.lexeme_props_getter = null_props
@ -94,8 +95,10 @@ def main(input_loc, freqs_dir, output_loc, n_jobs=2, skip_existing=False):
if not path.exists(output_path) or not skip_existing: if not path.exists(output_path) or not skip_existing:
tasks.append((input_path, output_path)) tasks.append((input_path, output_path))
if tasks:
parallelize(count_freqs, tasks, n_jobs) parallelize(count_freqs, tasks, n_jobs)
print "Merge"
merge_counts(outputs, output_loc) merge_counts(outputs, output_loc)