About this Abstract |
| Meeting |
MS&T25: Materials Science & Technology
|
| Symposium
|
Enhancing the Accessibility of Machine Learning-Enabled Experiments
|
| Presentation Title |
Ptychography Data Pipelines at the Advanced Photon Source |
| Author(s) |
Steven Henke, Hannah Parraga, Albert Vong, Oliver Hoidn, Nicholas Schwarz |
| On-Site Speaker (Planned) |
Steven Henke |
| Abstract Scope |
Ptychography is an imaging technique that is used to characterize the nanostructure of complex materials, devices, and biological samples. The recently upgraded Advanced Photon Source (APS) delivers a substantially brighter, more coherent beam, which particularly benefits ptychographic techniques. Accordingly, APS now features approximately ten specialized ptychography instruments, each capable of exceeding 10 TB of raw data production per day. To keep pace with this prolific data generation, we are developing (1) pipelines to leverage edge and leadership computing resources for automated processing during data acquisition, (2) a machine learning model called PtychoPINN to increase processing throughput, and (3) a user-friendly analysis application called Ptychodus to make the advanced algorithms and workflows portable across instruments and accessible to general users. These tools ensure that facility users can maximize the scientific output of their current experiments and pave the way for future innovations in experiment automation. |