Speaker
Description
Integration of high performance computing and machine learning / artificial intelligence into beamline operations is an important step for handling growing data volumes and to increase scientific productivity of users with large or complex datasets. Work to incorporate a new on-premise high performance computing cluster at the Canadian Light Source with beamline data processing and analysis will be presented. This includes AI-based, GPU-accelerated segmentation of CT data; automated MX crystal centering using object detection; HPC-acceleration of MX dataset processing, large multi-dataset, multi-modal spectromicroscopy imaging data analysis; automated interpolation of large unevenly spaced datasets; and automated reconstruction of CT data during beamtime.
| I plan to submit also conference proceedings | No |
|---|
