https://github.com/Dataherald/dataherald

Interact with your SQL database, Natural Language to SQL using LLMs
https://github.com/Dataherald/dataherald

ai database finetuning llm nl-to-sql rag sql text-to-sql

Last synced: 4 days ago
JSON representation

Interact with your SQL database, Natural Language to SQL using LLMs

• Host: GitHub
• URL: https://github.com/Dataherald/dataherald
• Owner: Dataherald
• License: apache-2.0
• Created: 2023-06-20T21:06:11.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2024-05-28T03:48:11.000Z (5 months ago)
• Last Synced: 2024-05-28T14:26:23.216Z (5 months ago)
• Topics: ai, database, finetuning, llm, nl-to-sql, rag, sql, text-to-sql
• Language: Python
• Homepage: https://dataherald.readthedocs.io/en/latest/
• Size: 4.33 MB
• Stars: 2,765
• Watchers: 24
• Forks: 183
• Open Issues: 7
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

• StarryDivineSky - Dataherald/dataherald
• AiTreasureBox - Dataherald/dataherald - 11-02_3347_0](https://img.shields.io/github/stars/Dataherald/dataherald.svg)|Interact with your SQL database, Natural Language to SQL using LLMs| (Repos)

README

        
# Dataherald monorepo



  





    Query your relational data in natural language. 






  

      

   |

  

      

   |

  

      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:

- Allow business users to get insights from the data warehouse without going through a data analyst

- Enable Q+A from your production DBs inside your SaaS application

- Create a ChatGPT plug-in from your proprietary data

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:

```bash

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](CONTRIBUTING.md).