boto3>=1.28.43 Brotli>=1.1.0 click>=8.1.7 Distance>=0.1.3 PyMuPDF>=1.24.5 loguru>=0.6.0 matplotlib>=3.8.3 numpy>=1.21.6 pandas>=1.3.5 fast-langdetect>=0.1.1 regex>=2023.12.25 termcolor>=2.4.0 wordninja>=2.0.0 scikit-learn>=1.0.2 nltk==3.8.1 s3pathlib>=2.1.1 pdfminer.six>=20231228 Levenshtein nltk rapidfuzz statistics openxlab #安装opendatalab pandas numpy matplotlib seaborn scipy scikit-learn tqdm htmltabletomd pypandoc