Speaker
Description
The aim of the ARCAFF (Active Region Classification and Flare Forecasting) project is to develop a beyond state-of-the-art flare forecasting system leveraging end-to-end deep learning models to significantly improve upon current flare forecasting capabilities. However, transitioning these models from research environments to robust, operational systems presents significant challenges, especially in ensuring their ongoing reliability and trustworthiness. This talk will focus on the MLOps (Machine Learning Operations) practices implemented within the ARCAFF project to address this. In particular, we will focus on reproducibility, detailing our methodologies for versioning data, code, and models, as well as utilising standardised environments to ensure consistency. Also, will touch upon our strategies for the continuous operation, timely updates, and vigilant monitoring of these models to maintain high performance and adapt to evolving solar conditions.
| Do you plan to attend in-person or online? | In-person |
|---|
