Skip to content

README.md

Latest commit

 

History

History
83 lines (52 loc) · 3.52 KB

README.md

File metadata and controls

83 lines (52 loc) · 3.52 KB

Dataherald Community App 🛠️

Welcome to the Dataherald Community App! This Streamlit-based web application allows users to interact with the Dataherald engine through a user-friendly interface, making it easier to work with your structured data. 🔥🔥🔥

Dataherald Logo

Table of Contents 📚

Features 🚀

  • Ask your Questions: 💡 Connect to various databases by providing their connection details and ask any questions that have on top of your mind from Dataherald's powerful NL-to-SQL engine.

  • Add Golden records: 💎 In order to boost the performance of the engine even higher, add verified Question/SQL pairs. These pairs are going to be used by the agent at inference time.

  • View and Remove Golden records: 📜 This application provides a user-friendly interface for viewing, searching, and removing Question/SQL pairs to control the agent's behavior.

  • Table Scanning: 🗃️ Scan tables in connected databases by specifying the table name. Scanned tables are used by the agent to generate answers.

  • Viewing Tables: 📊 View descriptions and details of tables within connected databases. Understand the structure and number of columns of each table easily.

Prerequisites ⚙️

Before running Dataherald, ensure that you have the following prerequisites:

  • Dataherald engine: 💻 Make sure you have Dataherald engine running since this application calls the endpoints provided by the Dataherald engine.

  • Python 3.x: 🐍 Make sure you have Python 3.x installed on your system. You can download it from the official Python website.

Installation 🚀

To get started with Dataherald, follow these steps:

  1. Clone the repository to your local machine:

    git clone https://github.com/Dataherald/streamlit-app.git
cd streamlit-app

  2. Create a virtual environment (optional but recommended):

    python -m venv venv
source venv/bin/activate  # On Windows, use `venv\Scripts\activate`

  3. Install required packages: 📦 This app relies on several Python packages. Install them by running the following command within the project directory:

    pip install -r requirements.txt

Usage 📈

To use Dataherald Community App and interact with databases, follow these steps:

Change the following code parts based on your current implementation of Dataherald engine:

DEFAULT_DATABASE = "RealEstate" # change this to your default database
HOST = st.sidebar.text_input("Engine URI", value="https://streamlit.dataherald.ai") #change the value to your localhost or server URI where your engine is deployed

Start the Dataherald Community App application:

streamlit run 🏠_Home.py

How it Works 🧐

If you want to know how the app works, you can take a look at the following figure:

Architecture

Contributions 🤝

We are extremely open to contributions, be it through new features, new pages, or improved documentation.

We are currently developing this repository, you can take a look at the roadmap we have for this app, and we will appreciate your contribution for the future works that we have. You can find the roadmap here.