A prototype written in Python to illustrate/demonstrate querying the Learning Path Index Dataset (see Kaggle Dataset and GitHub repo), with the help of the OpenAI GPT technology (InstructHPT model and embeddings model), Langchain and using Facebook's FAISS library.
The end-to-end process can be learnt by going through the code base as well as by observing the console logs when using both the Streamlit and the CLI versions.
- Python 3.8.x or above
- OpenAI API Key (see How to get an OpenAI API Key -- note it's may not be FREE anymore)
- Install dependencies from
requirements.txt
- Basic Command-line experience
- Basic git and GitHub experience
Copy the .env_template
to .env
in the current folder and then add your OpenAI API Key to .env
.
Please don't modify the .env_template
file.
pip install -r requirements.txt
python main.py
streamlit run main.py