🌸 Lilac

NEW: Try the Lilac hosted demo with pre-loaded datasets

👋 Welcome

Lilac is an open-source product that helps you analyze, structure, and clean unstructured data with AI.

Lilac can be used from our UI or from Python.

https://github.com/lilacai/lilac/assets/2294279/cb1378f8-92c1-4f2a-9524-ce5ddd8e0c53

💻 Install

To install Lilac on your machine:

pip install lilacai

You can also use Lilac with no installation by forking our public HuggingFace Spaces demo.

🔥 Getting started

Start the Lilac webserver from the CLI:

lilac start

Or start the Lilac webserver from Python:

import lilac as ll

ll.start_server()

This will open start a webserver at http://localhost:5432/.

📁 Documentation

Visit our website: lilacml.com

💻 Why Lilac?

Lilac is a visual tool and a Python API that helps you:

• Explore datasets with natural language (e.g. documents)
• Enrich your dataset with metadata (e.g. PII detection, profanity, text statistics, etc.)
• Conceptually search and tag your data (e.g. find paragraphs about injury)
• Remove unwanted or problematic data based on your own criteria
• Analyze patterns in your data

Lilac runs completely on device using powerful open-source LLM technologies.

💬 Contact

For bugs and feature requests, please file an issue on GitHub.

For general questions, please visit our Discord.