aiohttp==3.9.3
aiosignal==1.3.1
antlr4-python3-runtime==4.9.3
attrs==23.2.0
boto3==1.34.36
certifi==2024.2.2
cfgv==3.4.0
charset-normalizer==3.3.2
colorama==0.4.4
courlan==1.0.0
datasets==2.16.1
dateparser==1.2.0
Faker==23.0.0
fasttext-wheel==0.9.2
fsspec==2023.10.0
html2text
ipykernel>=6.29.1
ipython>=8.21.0
numpy==1.26.4
omegaconf==2.3.0
pandas==2.2.0
py4j==0.10.9.7
pyarrow==15.0.0
pyarrow-hotfix==0.6
pyspark==3.5.0
pytz==2024.1
regex==2023.12.25
requests==2.31.0
scipy==1.12.0
tqdm==4.66.1
trafilatura==1.7.0
