metadata

title: Scrum expert
emoji: 📚
colorFrom: purple
colorTo: blue
sdk: streamlit
sdk_version: 1.42.0
app_file: app.py
pinned: false
short_description: Scrum expert

Introduction

This is a RAG showcase easily adaptable for any set of documents (mainly pdf, docx, txt, csv).

How to run it locally ?

Clone the git repository
Replace the documents in ./data by your documents
Customize the constants at the beginning of app.py
Create a .streamlit directory
Create a .streamlit/secrets.toml file : openai_key="your-akash-api-key" (get your free key here : https://chatapi.akash.network/ > Get Started)
With .venv activated : pip install -r requirements.txt
Then python -m streamlit run app.py

Note : Every time you change the embedding model, it's necessary to delete the "storage" directory to rebuild the local vector db

How to run it on a new HuggingFace Space ?

When it runs locally, just commit and push to a new HuggingFace Space. You need to fill your Akash api key as a Secret in the "Settings > Variables and secrets" section of your space.