LangChainApps is a repo to store LLM experiments conducted with LangChain.
A Retrieval Augmented Generation python notebook which outlines the process of:
- Chunking and vectorizing the October 2023 United States Artificial Intelligence Executive Order
- Storing the vector embeddings locally in a Weaviate vector db
- Using an OpenAI API Key / GPT 3.5 Turbo to answer questions about the EO
A CLI python script which:
- Leverages an Open Source local LLM*, therefore no OpenAI API Key is necessary (thus no cost)
- Produces custom jokes on the fly
*7B parameter LLM is recommended for this type of exercise
🔧 Requirements include:
- argparse
- langchain
- weaviate
⚙️ To run the notebook, you will need to populate default.env with your OpenAI API key and rename it to .env
🐜 Function logging using @decorators allows for debugging