README.md

Weaviate Hybird, Reranker example

🎯 Overview

This is a simple demo of how one can run Weaviate vectorizer, reranker and generative Ollama modules and presents how one can perform meaningful searches.

📦 Requirements

In order to be able to create Weaviate one needs at least:

1. Docker
2. Python3

💡 Running

In order to run the setup one needs to issue:

docker compose up -d

Please note that below operations take some time to succeed.

Pull Aya-Expanse model. It's an open source multi linugal large language model from Cohere. Please note that you need at least 5GB of free space on your disk.

docker exec -i generative_ollama ollama pull aya-expanse:8b

In order to run notebooks, it's advised to setup a venv for a project.

python3 -m venv .venv
source .venv/bin/activate
pip3 install -r requirements.txt

or using uv:

uv venv --python 3.13
source .venv/bin/activate
uv pip install -r requirements.txt

📖 How to use notebooks

Prior running the examples please run import notebook:

1. Import data: (0-import.ipynb)

Examples:

1. Reranker example: (1-reranker.ipynb)
2. Hybrid example: (2-hybrid.ipynb)
3. RAG example: (3-rag.ipynb)

🔗 Useful links

• Dataset used in notebooks
• Vector similarity search
• Keyword search
• Hybrid search
• Reranking
• Generative AI / RAG