A framework for building an AutoML System. The toolkit is designed to be modular and extensible, allowing you to easily swap out components and integrate your own. The toolkit is designed to be used in a variety of different ways, whether for research purposes, building your own AutoML Tool or educational purposes.
We focus on building complex parametrized pipelines easily, providing tools to optimize these pipeline parameters and lastly, providing tools to schedule compute tasks on a variety of different compute backends, without the need to refactor everything, once you swap out any one of these.
The goal of this toolkit is to drive innovation for AutoML Systems by:
Allowing concise research artifacts that can study different design decisions in AutoML.
Enabling simple prototypes to scale to the compute you have available.
Providing a framework for building real and robust AutoML Systems that are extensible by design.
Institute of Medical Biometry and Statistics,
Faculty of Medicine and Medical Center –
University of Freiburg
