search

Dataherald / dataherald

Interact with your SQL database, Natural Language to SQL using LLMs
https://dataherald.readthedocs.io/en/latest/
Apache License 2.0
3.36k stars 234 forks source link
ai database finetuning llm nl-to-sql rag sql text-to-sql

readme

Dataherald monorepo

Dataherald logo

Query your relational data in natural language.

Discord | License | Docs | Homepage

Dataherald is a natural language-to-SQL engine built for enterprise-level question answering over relational data. It allows you to set up an API from your database that can answer questions in plain English. You can use Dataherald to:

This repository contains four components under /services which can be used together to set up an end-to-end Dataherald deployment:

  1. Engine: The core natural language-to-SQL engine. If you would like to use the dataherald API without users or authentication, running the engine will suffice.
  2. Enterprise: The application API layer which adds authentication, organizations and users, and other business logic to Dataherald.
  3. Admin-console: The front-end component of Dataherald which allows a GUI for configuration and observability. You will need to run both engine and enterprise for the admin-console to work.
  4. Slackbot: A slackbot which allows users from a slack channel to interact with dataherald. Requires both engine and enterprise to run.

For more information on each component, please take a look at their README.md files.

Running locally

Each component in the /services directory has its own docker-compose.yml file. To set up the environment, follow these steps:

  1. Set Environment Variables: Each service requires specific environment variables. Refer to the .env.example file in each service directory and create a .env file with the necessary values.

    For the Next.js front-end app is .env.local

  2. Run Services: You can run all the services using a single script located in the root directory. This script creates a common Docker network and runs each service in detached mode.

Run the script to start all services:

sh docker-run.sh

Contributing

As an open-source project in a rapidly developing field, we are open to contributions, whether it be in the form of a new feature, improved infrastructure, or better documentation.

For detailed information on how to contribute, see here.