explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2024-09-21 19:39:13 +03:00
Code Issues Packages Projects Releases Wiki Activity

* Fix download script

Browse Source
This commit is contained in:
Matthew Honnibal 2015-01-25 23:01:10 +11:00
parent 8cd5a91063
commit 7f87716cf7
1 changed files with 10 additions and 30 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

40
spacy/en/download.py
View File

@ -2,53 +2,33 @@ from os import path
import os import os
import tarfile import tarfile
import shutil import shutil
import requests import wget
PARSER_URL = 'http://s3-us-west-1.amazonaws.com/media.spacynlp.com/en.tgz' PARSER_URL = 'http://s3-us-west-1.amazonaws.com/media.spacynlp.com/en_deps-0.30.tgz'
DEP_VECTORS_URL = 'http://u.cs.biu.ac.il/~yogo/data/syntemb/deps.words.bz2'
DEP_VECTORS_URL = 'http://s3-us-west-1.amazonaws.com/media.spacynlp.com/vec.bin'
DEST_DIR = path.join(path.dirname(__file__), 'data') DEST_DIR = path.join(path.dirname(__file__), 'data')
def download_file(url): def download_file(url, out):
local_filename = url.split('/')[-1] wget.download(url, out=out)
return path.join(DEST_DIR, local_filename)
# NOTE the stream=True parameter
r = requests.get(url, stream=True)
print "Download %s" % url
i = 0
with open(local_filename, 'wb') as f:
for chunk in r.iter_content(chunk_size=1024):
if chunk: # filter out keep-alive new chunks
f.write(chunk)
f.flush()
print i
i += 1
return local_filename
def install_parser_model(url, dest_dir): def install_parser_model(url, dest_dir):
if not os.path.exists(dest_dir): download_file(url, dest_dir)
os.mkdir(dest_dir) t = tarfile.open(path.join(DEST_DIR, 'en_deps-0.30.tgz'), mode=":gz")
assert not path.exists(path.join(dest_dir, 'en'))
filename = download_file(url)
t = tarfile.open(filename, mode=":gz")
t.extractall(dest_dir) t.extractall(dest_dir)
shutil.move(path.join(dest_dir, 'en', 'deps', 'model'), dest_dir)
shutil.move(path.join(dest_dir, 'en', 'deps', 'config.json'), dest_dir)
shutil.rmtree(path.join(dest_dir, 'en'))
def install_dep_vectors(url, dest_dir): def install_dep_vectors(url, dest_dir):
if not os.path.exists(dest_dir): if not os.path.exists(dest_dir):
os.mkdir(dest_dir) os.mkdir(dest_dir)
filename = download_file(url) filename = download_file(url, dest_dir)
shutil.move(filename, path.join(dest_dir, 'vec.bz2'))
def main(): def main():
#install_parser_model(PARSER_URL, path.join(DEST_DIR, 'deps')) install_parser_model(PARSER_URL, DEST_DIR)
install_dep_vectors(DEP_VECTORS_URL, path.join(DEST_DIR, 'vocab')) install_dep_vectors(DEP_VECTORS_URL, path.join(DEST_DIR, 'vocab'))