This Repo contains a simple script that demonstrate the use of RAG which can answer user's questions regarding the NIST SP 800-53 Rev. 5 document.
The script is based on the DeepLearning.AI courses: LangChain: Chat with Your Data and Open Source Models with Hugging Face.
The script use the NIST SP 800-53 pdf document and build the embeddings using Hugging face's all-MiniLM-L6-v2 model. Based on the embeddings, the script looks for the relevant context related to the question and use OpenAI gpt-3.5-turbo model to generate an answer.
Result:
