Albert "Evaluation General One" is a high-level service designed to perform evaluations for Etalab. This project provides an API to evaluate [LLM] models and an interface to navigate datasets, models, metrics and experiments.
For guidance on how to use this project, please refer to the following resources:
- the demo notebooks: notebooks/
- The api reference: https://eg1.etalab.gouv.fr/redoc
- The user-friendly interface: https://eg1.etalab.gouv.fr/
You can open issues for bugs you've found or features you think are missing. You can also submit pull requests to this repository. To get started, take a look at CONTRIBUTING.md.
To be defined.