search

test-app-organization / test-repo-app

MIT License
0 stars 0 forks source link

Does it support multi modal RAG queries? #55

Open Sheth-G opened 7 months ago

Sheth-G commented 7 months ago

RAG is Retrieval Augmented Generation. For example. if i pass a picture, will it find a similar?

torvalds-dev commented 7 months ago

torvalds.dev is analyzing the ticket

torvalds-dev commented 7 months ago

The task is to understand the issue: "Does it support multi-modal RAG queries?" and suggest relevant actions that should be taken, as well as provide the names of files that are relevant to the issue.

RAG stands for Retrieval Augmented Generation. It's a method used in Natural Language Processing (NLP) to generate responses to queries by retrieving relevant information from a database or knowledge graph.

The question is whether the current system supports multi-modal RAG queries. Multi-modal queries are those that involve more than one type of data, such as text and images. For example, if a picture is passed to the system, will it find a similar one?

The files provided are related to the Llama Index, a system for retrieving and synthesizing responses to queries. The no_text.py file defines a class NoText that returns an empty string for any query. The retrievers.py file defines classes and methods for retrieving information from a knowledge graph. The rag.py file defines a class RAG that uses a query engine to generate responses to queries.

Plan:

  1. Analyze the provided files to understand how the system currently handles queries.
  2. Determine whether the system supports multi-modal queries.
  3. If it does not, suggest actions to add this functionality. If it does, suggest actions to improve or optimize this functionality.