Testing out how RAG would work with locally hosted models.
-
Install Ollama
-
Run the following command to create python virtual environment and start it:
python3 -m venv venv source venv/bin/activate
-
Run the following command to install dependencies:
pip install -r requirements.txt
-
Set environment variables inside an
.env
file:RSS_FEED_PATH
: The path to RSS feed you wan't to use as source for your documents e.f. NYTCOLLECTION_NAME
: The name for ChromaDB collection. You can choose one that is most to your liking e.g.docs
EMBEDDING_MODEL
: Name of the embedding model e.g.mxbai-embed-large
GENERATIVE_MODEL
: Name of the generative model e.g.phi3
-
Run:
python src/main.py
-
When you want to end the conversation then just write
bye
Streamlit UI coming maybe soonish..