This repository contains a local RAG (Retrieval-Augmented Generation) application written in Python. It processes documents to convert them into vectors, enabling efficient retrieval of relevant information to answer user queries without sending entire documents repeatedly.
From the README
Local RAG Document Assistant What is this? This is a local RAG (Retrieval-Augmented Generation) application. You give it documents such as PDFs or text files. You can then ask questions about those documents. Instead of asking an AI model to answer from memory, the system first searches the documents for relevant information. Simple flow Why use RAG? Suppose you have a 100-page document. Instead of sending the entire document every time: 1. The document is processed once. 2. Its sections are converted into vectors. 3. A question is converted into a vector. 4. The closest sections are found. 5. Those sections are given to the language model. 6. The model generates an answer using...
Skills
- Python
