Speaker
Oliver Humphries
(Eur.XFEL (European XFEL))
Description
High rep-rate data requires high rep-rate solutions, where human-in-the-loop decision making no longer sufficient to manage many experiments efficiently. Simultaneously, the requirement for high-level, (near-) online analysis to manage experiments is often explicitly at odds with their goal: to do something new where you don't know exactly what you'll see. I'll walk through how HED have built from early, often user-provided notebooks/codebases, toward (more) robust, adaptable analysis deployed across many experiments - making use of DAMNIT and extra-metro frameworks at European XFEL - focusing on the example of Talbot phase imaging.
