https://github.com/zeno-ml/zeno-hub
AI Evaluation Platform
https://github.com/zeno-ml/zeno-hub
ai evaluation machine-learning visualization
Last synced: about 1 year ago
JSON representation
AI Evaluation Platform
- Host: GitHub
- URL: https://github.com/zeno-ml/zeno-hub
- Owner: zeno-ml
- License: mit
- Created: 2023-07-14T18:09:38.000Z (almost 3 years ago)
- Default Branch: main
- Last Pushed: 2025-01-04T06:06:05.000Z (over 1 year ago)
- Last Synced: 2025-01-04T06:20:16.194Z (over 1 year ago)
- Topics: ai, evaluation, machine-learning, visualization
- Language: CSS
- Homepage: https://hub.zenoml.com
- Size: 3.57 MB
- Stars: 45
- Watchers: 2
- Forks: 10
- Open Issues: 11
-
Metadata Files:
- Readme: README.md
- License: LICENSE.txt
Awesome Lists containing this project
- AwesomeResponsibleAI - Zeno Hub
README
# Zeno Hub

[](https://dl.acm.org/doi/10.1145/3544548.3581268)
[](https://lbesson.mit-license.org/)
[](https://discord.gg/km62pDKAkE)
[](https://hub.zenoml.com)
[Zeno Hub](https://hub.zenoml.com) is an interactive AI evaluation platform.
With Zeno, you can go from exploring your data and model outputs to creating rich visualizations and reports. Learn more about Zeno at [zenoml.com](https://zenoml.com)
## Self-Hosting
We recommend you use the [hosted version of Zeno](https://hub.zenoml.com), but you can run Zeno locally with `docker-compose`:
1. [Send us an email](hello@zenoml.com) to get the environment variables to put in `/frontend/.env`
2. Run `docker-compose up -d`
## Development
Zeno Hub consists of three Docker containers:
- **[SvelteKit Frontend](./frontend/)** - SvelteKit application for rendering the frontend interface. Actually runs both a backend and frontend.
- **[FastAPI Backend](./backend/)** - FastAPI Python backend that routes requests from the frontend to database and does processing such as data upload, error discovery algorithms, etc.
- **PostgreSQL Database** - We store all application _and_ analysis data in a PostgreSQL database.
These services are orchestrated using `docker-compose` and deployed using Terraform and AWS ECS containers.
Please see [DEVELOPMENT.md](./DEVELOPMENT.md) for more information on how to install, run, and develop Zeno locally.