We introduce a web-based, easy-to-use interface for designing machine learning pipelines that generates PHOTON Python code from user interaction while implicitly conveying machine learning knowledge and code comprehension.
With the Wizard’s intuitive web-based approach to model creation, we remove the need to learn coding so that you can focus on your machine learning project.
The PHOTON Wizard automatically adapts to your design choices by filtering suitable methods and suggesting default parameters. Greatly promoting code comprehension, syntax is generated while the user sets up his/her pipeline.
The Wizard generates native PHOTON Python code, including hyperparameter optimization and performance estimation. The model can be saved and shared in a standardized format to enable prediction across sites and in different application scenarios.