README.md

metadata

title: DataVision AI
emoji: 📊
colorFrom: blue
colorTo: indigo
sdk: docker
app_port: 7860
pinned: false

DataVision AI

Autonomous analytics, business intelligence, AutoML, computer vision, and team collaboration in one platform.

Live application · Product user guide (PDF)

What DataVision provides

• Interactive dashboards with slicers, KPI cards, business charts, AI explanations, exports, and themes.
• AI Analyst for grounded questions about uploaded datasets.
• AutoML, predictions, forecasts, reports, vector search, and data pipelines.
• Computer Vision for detection, classification, segmentation, pose estimation, and OCR.
• Persistent collaboration with channels, replies, reactions, pinning, and message deletion.
• Developer API keys, webhooks, usage reporting, and embed tools.

Run locally

Requirements: Python 3.11+, Node.js 20+, and PostgreSQL for persistent production data.

# Backend
cd backend
python -m venv .venv
.venv\Scripts\activate
pip install -r requirements.txt
python main.py

# Frontend (in a second terminal)
cd frontend
npm install
npm run dev

The frontend opens at http://localhost:5173.

Configuration

Create backend/.env with your production settings. Do not commit secrets.

DATABASE_URL=postgresql+asyncpg://user:password@host:5432/datavision
GROQ_API_KEY=your_key
JWT_SECRET=a_long_random_secret
ENVIRONMENT=development
CORS_ORIGINS=http://localhost:5173

Deployment

Pushes to main run the GitHub Actions workflows in .github/workflows/. Configure these GitHub secrets before deployment:

• HF_TOKEN (Hugging Face write token)
• Optional HF_SPACE (defaults to datavision-ai/Datavision)

Set DATABASE_URL, GROQ_API_KEY, and JWT_SECRET in your Hugging Face Space settings. See HUGGINGFACE_DEPLOYMENT.md for the checklist.

User guide

The updated guide is available here: DataVision AI Product User Guide.

Validation

cd frontend
npm run build

cd backend
python -m py_compile main.py api/v1/endpoints/collaboration.py