The Part1_Initial_prompt_test.py is one of the first scripts I mention in the AI-Residency first blog post! It explores some basic capabilities of LangChain and how Prompting Architectures can be built!
The Evaluation Folder contains the remaining relevant files mentioned in the first part!
The Assistant folder contains the streamlit_app.py, the requirements.txt for it and an example.env! This python file contains all code related to building the AI assistant (GPT-3.5-turbo based) above the main function.
From the main function onwards, the code aims to incorparate the calling of the LLM assistant into the streamlit application.
The streamlit app requires a .env file that contains all the relevant keys! You must also set a name, username and password in streamlit_app.py code in order to be able to log in!
To use the application you must enter the following in the terminal:
streamlit run streamlit_app.py