def main():
args = setup_args()
glove_base_url = "http://nlp.stanford.edu/data/"
glove_filename = "glove.6B.zip"
print "\nDownloading wordvecs to {}".format(args.download_dir)
if not os.path.exists(args.download_dir):
os.makedirs(args.download_dir)
maybe_download(glove_base_url, glove_filename, args.download_dir, 862182613L)
glove_zip_ref = zipfile.ZipFile(os.path.join(args.download_dir, glove_filename), 'r')
glove_zip_ref.extractall(args.download_dir)
glove_zip_ref.close()
def main():
args = setup_args()
glove_base_url = "http://nlp.stanford.edu/data/"
glove_filename = "glove.6B.zip"
print("\nDownloading wordvecs to {}".format(args.download_dir))
if not os.path.exists(args.download_dir):
os.makedirs(args.download_dir)
maybe_download(glove_base_url, glove_filename, args.download_dir, 862182613)
glove_zip_ref = zipfile.ZipFile(os.path.join(args.download_dir, glove_filename), 'r')
glove_zip_ref.extractall(args.download_dir)
glove_zip_ref.close()
def main():
args = setup_args()
glove_base_url = "http://nlp.stanford.edu/data/"
glove_filename = "glove.840B.300d.zip"
if not os.path.exists(args.data_dir):
os.makedirs(args.data_dir, exist_ok=True)
maybe_download(glove_base_url, glove_filename, args.data_dir)
glove_path = os.path.join(args.data_dir, glove_filename)
print("Unzipping {}...".format(glove_path))
glove_zip_ref = zipfile.ZipFile(glove_path, 'r')
glove_zip_ref.extractall(args.data_dir)
glove_zip_ref.close()
if __name__ == '__main__':
parser = argparse.ArgumentParser()
parser.add_argument(
"--glove_source",
choices=['wiki', 'crawl_ci', 'crawl_cs'],
required=True) # added for use with different glove sources
args = parser.parse_args()
glove_base_url = "http://nlp.stanford.edu/data/"
if args.glove_source == 'wiki':
glove_filename = "glove.6B.zip"
elif args.glove_source == 'crawl_ci':
glove_filename = "glove.42B.300d.zip"
elif args.glove_source == 'crawl_cs':
glove_filename = "glove.840B.300d.zip"
prefix = os.path.join("download", "dwr")
print("Storing datasets in {}".format(prefix))
if not os.path.exists(prefix):
os.makedirs(prefix)
glove_zip = maybe_download(glove_base_url, glove_filename, prefix,
862182613)
glove_zip_ref = zipfile.ZipFile(os.path.join(prefix, glove_filename), 'r')
glove_zip_ref.extractall(prefix)
glove_zip_ref.close()
import argparse
import os
from squad_preprocess import maybe_download
def setup_args():
parser = argparse.ArgumentParser()
parser.add_argument(
"--download_dir",
required=True) # where to put the downloaded glove files
return parser.parse_args()
if __name__ == '__main__':
args = setup_args()
glove_base_url = "http://nlp.stanford.edu/data/"
glove_filename = "glove.6B.zip"
print("\nDownloading wordvecs to {}".format(args.download_dir))
if not os.path.exists(args.download_dir):
os.makedirs(args.download_dir)
# if not os.path.exists(os.path.join(prefix, filename)):
maybe_download(glove_base_url, glove_filename, args.download_dir,
862182613) # instead of 862182613L (python2)
glove_zip_ref = zipfile.ZipFile(
os.path.join(args.download_dir, glove_filename), 'r')
glove_zip_ref.extractall(args.download_dir)
glove_zip_ref.close()
