FrancyJGLisboa / OI_Memory

Making a way to have LLMs autonomously construct, query and update an external database from raw data.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

OI_Memory

Making a way to have LLMs autonomously construct, query and update an external database from raw data.

Day 1

  • Measure performance of system:
    • Create library (Pdfs, textbooks, documentation, code etc)
    • Create queries corresponding to a chunk of text (To test performance of external database)
  • Write cloud functions to extract text from raw data like:
    • PDFs
    • Codebases on github
    • Documentation on web
  • Implement system:
    • Create binary tree with LLMs using sumarization(Constructing tree) and comparison(Traversing tree) first
    • Add features:
      • Try optimizing for tokens, aka, not strictly a binary tree
      • Allow for clone nodes, implement cleanup operation

About

Making a way to have LLMs autonomously construct, query and update an external database from raw data.

License:MIT License


Languages

Language:Python 100.0%