Make a JFK Archives GPT

TL;DR

Getting all pdf from the JFK Assassination Records to train a GPT to be an expert on this case.

First clone this repo and install dependancies

pip install --upgrade pip
pip install pandas requests openpyxl  pymupdf pytesseract

brew install tesseract

They are listed in the jfk.xls file

python3 download.py

Once all PDF are collected, start Tesseract to extract text from PDF into 1 Mb txt files (openai doesn't like larger txt files)

python3 extract.py

All the code was written (after many try) with the help of GPT4