How to get started using LunaBrain.
Go is a very easy language to learn. It’s also very fast, efficient, and easy to scale.
Python is ubiquitous in ML. There is a Python service which serves as a bridge between the Go backend and the ML models.
Clone the repository
git clone https://github.com/lunabrain-ai/lunabrain.git cd lunabrain
The fastest way to bring up LunaBrain is with
Run the backend
If you want to roll up your sleves and dig around, you can bring up the services manually.
Note: These might fall out of date quickly! We are working on a better local dev experience with tilt.
go run cmd/main.go content serve
What can I configure?
config/lunabrain/config.yaml file and add the following config (changing TODO with the relevant details):
openai: api_key: TODO api: local: true scrape: client: chrome fallback: true use_cache: false python: host: 'lunabrain-python:50051' normalize: url: domain_content: true youtube: api_key: TODO discord: application_id: TODO token: TODO publish: discord: enabled: false channel_id: TODO
Start the Python service
There is a Python service which serves as a bridge between the Go backend and the ML models. It is required to run the backend. The code is currently a work in progress, but you can find the latest details here.
cd python python -m venv env source env/bin/activate pip install -r requirements.txt LUNABRAIN_DIR=data/lunabrain python start.py
If you want to explore what this service can do, you can use grpcui:
grpcui -plaintext localhost:50051