I didn't like opensource versions, so I did my own
python3 -m venv venv
source venv/bin/activate
python -m pip install pip setuptools wheel
python -m pip install -e .
Enter your OPEN AI API KEY top config.py
file
python main.py --filepath=PDF_FILE_PATH
If input text is long than the maximum number of tokens the model can handle, text is slitted into multiple chucks and the results are concatenated. Final text is saved to cache
folder (unless configured otherwise) with postfix _summary