NeurIPS_paper_downloader.py

# made by ryul99 at github

from bs4 import BeautifulSoup
import requests
from tqdm import tqdm

paper_index_page = 'whole url of NeaurIPS paper list page'
print('Crawling Paper List...')
req = requests.get(paper_index_page)
html = req.text
soup = BeautifulSoup(html, 'html.parser')
links = [obj.get('href') for obj in soup.find_all('a') if 'paper/' in obj.get('href')]
for link in tqdm(links):
    paper_link = 'https://papers.nips.cc/' + link + '.pdf'
    paper = requests.get(paper_link)
    paper_name = link.split('/')[-1] + '.pdf'
    with open(paper_name, 'wb') as f:
        for chunk in paper.iter_content(chunk_size=2000):
            f.write(chunk)