search

mt7180 / quaigle

RAG-based LLM application: project to explore different llm-frameworks like llamaindex, marvin and langchain, other used frameworks: streamlit, fastapi
1 stars 1 forks source link

6 detail backend api for qa on text files #14

Closed Zaubeerer closed 1 year ago

mt7180 commented 1 year ago

Hey @Robin, I got my LlamaIndex concept to work and pushed it to this branch. You can find it in script.py, next step after merging would be to implement it into fastapi_app.py ...

mt7180 commented 1 year ago

Hey @Zaubeerer , I think you didn't receive my last comment. Can we merge this into main? I would also like to merge branch #16 into it, and I would start with #17 , should I open 2 more pull requests now?

mt7180 commented 1 year ago

Well, I know it is a little bit confusing, and next time I will do better, but the "new" stuff here is the script.py code where I used Llamaindex. In #16 I have cleaned up the fastapi_app.py file and implemented the first route WITHOUT marvin ... so even if we would merge marvin in here, with the next pull request it will be deleted again ... sorry for the confusion ...

mt7180 commented 1 year ago

Probably it would make sense to ignore this pull request and just merge #16 into main, since the relevant code is also in there and it is not mixed with marvin ...

Zaubeerer commented 1 year ago

I would still merge this code, but put it in a POC folder.

So, in interviews you can demonstrate what you did and why it did not work, which is also great.

mt7180 commented 1 year ago

As suggested I have put the code to the PoC Marvin folder. As next step I would merge this pull request and thereafter the next one #19 ...