Parsing penalties from F1 documentation using LLM
Opened this issue · 2 comments
Hello,
I created a repo with a fairly different approach for extracting information from the F1 documentation https://github.com/marcll/f1-fia-doc-parser
I am extracting the info using a LLM instead of parsing the documents, in some cases may be a good enough and more adaptable solution especially when handling document format changes. Feel free to close the issue and check and draw any code or inspiration from it.
Thanks
Thank you! This is really amazing. I will update the README page here and put a link to your repo. We've been thinking about, but did not make much progress, digitalising the "less-data" PDFs, e.g. penalty, type compound, director notes, etc. I will look into details when I get time.
Awesome @harningle :-) hope it can be helpful!