Skip to content

BERT-PyT subprocess for bzip in wikidownloader #180

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
szmigacz merged 3 commits into from
Aug 29, 2019
Merged

BERT-PyT subprocess for bzip in wikidownloader #180

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
8474c12
Removing unnecessary subprocess.communicate calls
nvcforster Aug 22, 2019
7df14d2
Updating Bookscorpus downloader to require less memory
nvcforster Aug 22, 2019
5f15123
Renaming variable
nvcforster Aug 28, 2019
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 0 additions & 1 deletion PyTorch/LanguageModeling/BERT/data/BooksDownloader.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -13,4 +13,3 @@ def download(self):
bookscorpus_download_command += ' ' + self.save_path + '/bookscorpus'
bookscorpus_download_command += ' --trash-bad-count'
bookscorpus_download_process = subprocess.run(bookscorpus_download_command, shell=True, check=True)
bookscorpus_download_process.communicate()
20 changes: 5 additions & 15 deletions PyTorch/LanguageModeling/BERT/data/WikiDownloader.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -3,6 +3,7 @@
import bz2
import os
import urllib.request
import subprocess
import sys

class WikiDownloader:
Expand All @@ -27,32 +28,21 @@ def __init__(self, language, save_path):
def download(self):
if self.language in self.download_urls:
url = self.download_urls[self.language]
file = self.output_files[self.language]
filename = self.output_files[self.language]

print('Downloading:', url)
if os.path.isfile(self.save_path + '/' + file):
if os.path.isfile(self.save_path + '/' + filename):
print('** Download file already exists, skipping download')
else:
response = urllib.request.urlopen(url)
with open(self.save_path + '/' + file, "wb") as handle:
with open(self.save_path + '/' + filename, "wb") as handle:
handle.write(response.read())

# Always unzipping since this is relatively fast and will overwrite
print('Unzipping:', self.output_files[self.language])
#with open(self.save_path + '/' + file, mode='rb', buffering=131072) as f:
# it = iter(lambda: f.read(131072), b'')
# self.decompression(it, sys.stdout.buffer)

zip = bz2.BZ2File(self.save_path + '/' + file)
open(self.save_path + '/wikicorpus_' + self.language + '.xml', mode='wb', buffering=131072).write(zip.read())
subprocess.run('bzip2 -dk ' + self.save_path + '/' + filename, shell=True, check=True)

else:
assert False, 'WikiDownloader not implemented for this language yet.'

def decompression(self, input, output):
decomp = bz2.BZ2Decompressor()

for chunk in input:
dc = decomp.decompress(chunk)
output.write(dc)

2 changes: 0 additions & 2 deletions PyTorch/LanguageModeling/BERT/data/bertPrep.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -70,7 +70,6 @@ def main(args):
wikiextractor_command = path_to_wikiextractor_in_container + ' ' + directory_structure['download'] + '/' + args.dataset + '/wikicorpus_en.xml ' + '-b 100M --processes ' + str(args.n_processes) + ' -o ' + directory_structure['extracted'] + '/' + args.dataset
print('WikiExtractor Command:', wikiextractor_command)
wikiextractor_process = subprocess.run(wikiextractor_command, shell=True, check=True)
#wikiextractor_process.communicate()

wiki_path = working_dir + '/' + directory_structure['extracted'] + '/wikicorpus_en'
output_filename = directory_structure['formatted'] + '/wikicorpus_en_one_article_per_line.txt'
Expand All @@ -84,7 +83,6 @@ def main(args):
wikiextractor_command = path_to_wikiextractor_in_container + ' ' + directory_structure['download'] + '/' + args.dataset + '/wikicorpus_zh.xml ' + '-b 100M --processes ' + str(args.n_processes) + ' -o ' + directory_structure['extracted'] + '/' + args.dataset
print('WikiExtractor Command:', wikiextractor_command)
wikiextractor_process = subprocess.run(wikiextractor_command, shell=True, check=True)
#wikiextractor_process.communicate()

wiki_path = working_dir + '/' + directory_structure['extracted'] + '/wikicorpus_zh'
output_filename = directory_structure['formatted'] + '/wikicorpus_zh_one_article_per_line.txt'
Expand Down