sprintup / rag

Build a retrieval augmented generated (RAG) model

Figuring out how to build a RAG model to submit to the below contest (submissions due June 17, 2024). The submission just has to use NVIDIA technologies and Langchain. They'd prefer you to use their endpoints but from what I gather, if you can prove how you use NVIDIA technologies, they'll consider your submission.

• How does RAG work in 60 seconds
• What is LangChain?
• https://www.nvidia.com/en-us/ai-data-science/generative-ai/developer-contest-with-langchain/
• https://developer.nvidia.com/generative-ai-agent-contest-registration/thank-you
• https://developer.nvidia.com/blog/generative-ai-agents-developer-contest-top-tips-for-getting-started
• https://github.com/NVIDIA/GenerativeAIExamples
• https://docs.nvidia.com/nemo/guardrails/user_guides/llm/nvidia_ai_endpoints/README.html

Build a Retrieval-Augmented Generation Chatbot in 5 Minutes

source code

Large Language Models: Application through Production

The above diagram is an explainer from databricks about how RAG databases work from the edx course. The include LLM files will only run in databricks.

2.7.1 Notebook Demo Part 1

• the source code for the notebook is found here

Misc Resources

• https://abvijaykumar.medium.com/ollama-build-a-chatbot-with-langchain-ollama-deploy-on-docker-5dfcfd140363
• What is a Vector Database?
• Langchain docs
• LM Studio
• Hugging Face
• Databricks
• Using generate_events_async and Streaming
• NVIDIA LLM Developer Day Videos
  • The Fast Path to Developing with LLMs
    • 
  • Tailoring LLMs to Your Use Case
    • 
  • Large Language Models and Generative AI for Life Sciences
  • Running Your Own LLM
    • 
  • Reinventing the Complete Cybersecurity Stack With AI Language Models
    • 
  • Technical Ask-the-Experts

Install

• python -m venv venv_rag
• source vevn_rag/bin/activate
• pip install -r requirements.txt

or

pip install streamlit
pip install langchain_nvidia_ai_endpoints
pip install faiss-cpu
pip install langchain
pip install -U langchain-community
pip install unstructured

Start

• source .env
• source venv_rag/bin/activate
• streamlit run main.py