Midterm Certification Challenge: RAG
In this repo, you'll find a pretty entertaining RAG application, using Langchain, OpenAI, text-3-embedding small, GPT-3.5-turbo, the FAISS vectorstore, and an extra component: visibility tooling with WandB.
⚙️The colab link to the code is found and (will also be included in this repo) here.
🫂The video walkthrough can be found here.
- Data: NVIDIA 10-k Filings
- Model: OpenAI text-3-embedding small, GPT-3.5-turbo
- Tooling: LangChain or LlamaIndex (option to choose)
- Vector Store: FAISS
- Additional Component: Add one of the following: 1) visibility with WandB OR 2) evaluation with RAGAS
Evaluate your answers to the following questions:
- "Who is the E-VP, Operations - and how old are they?"
- "What is the gross carrying amount of Total Amortizable Intangible Assets for Jan 29, 2023?"