The RAG App workshop for Momento Vector Index

Momento Vector Index example for handson using Python, LangChain, OpenAI and Streamlit.

This repository is originally made for ServerlessDays Tokyo 2023 Momento Hands-on Workshop

Features

• RAG(Retrieval Augmented Generation) web app.
• Getting embeddings(vector data) from local file using OpenAI Embeddings API.
• Hosted the vector data to Momento Vector Index and retrieve from it.
• Cached the Q&A histories to Momento Cache.
• Answer generated using OpenAI.
• UI was created using Streamlit.

Getting Started

1. Copy .env.template to .env.
2. Get and fill the API key to MOMENTO_AUTH_TOKEN and OPENAI_API_KEY in .env.
3. Change the source data in data/sample.txt.
4. Create virtual environment

curl https://pyenv.run | bash
echo 'export PYENV_ROOT="$HOME/.pyenv"' >> ~/.bashrc
echo 'command -v pyenv >/dev/null || export PATH="$PYENV_ROOT/bin:$PATH"' >> ~/.bashrc
echo 'eval "$(pyenv init -)"' >> ~/.bashrc
exec "$SHELL"
pyenv --version
sudo yum remove -y openssl-devel
sudo yum install -y openssl11-devel bzip2-devel xz-devel
pyenv install 3.10
pyenv local 3.10
rm -rf .venv
python3 -m venv .venv
. .venv/bin/activate

5. Install the libraries.

$ pip install -r requirements.txt

6. Generate vector data

python insert.py

7. Run streamlit, Access to http://localhost:8051 and ask sometiong about your data

streamlit run app.py

8. Delete vector data

python delete.py

Feedback

You are welcome to discuss it on our discord.