Personal Rag Dev Log
TODOs:
- Properly add documents to chroma collection
- Setup the RAG
- Check RAG with RAGAS
- Create Knowledge base to add to chroma
May 25, 2025
May 24, 2025
- Create a test personal knowledge base to test the quality of the DB and RAG.
- ?? What is the ideal chunk size for documents in a Vector DB and for RAG.
- The tool RAGAS link can evaluate the Model
- Types of RAG Queries Single-Hop, and Multi-Hop
- Single-Hop: A single-hop query is a straightforward question that requires retrieving information from a single document or source to provide a relevant answer. It involves only one step to arrive at the answer.
- Multi-Hop: A multi-hop query involves multiple steps of reasoning, requiring information from two or more sources. The system must retrieve information from various documents and connect the dots to generate an accurate answer.
- Testing RAG: ragas docs
- TESING NEW âž• 2025-05-24
May 23,2025
Built out a Test Chroma DB
- have small scripts for CRUD actions
- Used Streamlit to have visualization of the DB
- Technically the Python scripts made after setting up the Chroma DB use RAG to reference the Vector DB. (Needs more testing)
When building the DB need to collect and properly label notes and docs into MarkDown with Metadata Which the llm with use to reference.
Also limiting Docs to 250-500 words for easier consumption
looked at Dolphin-mcp-cli didn’t quite work out of the box could help out when building my own to see a architecture that supposedly works.
- TESTING âž• 2025-05-24