Speaker
Description
Authors: Piero Gasparotto, Luis Barba, Hans-Christian Stadler,
Greta Assmann, Henrique Mendonça, Alun W. Ashton, Markus Janousch,
Filip Leonarski and Benjamı́n Béjar
TORO (TOrch-powered Robust Optimization) [1] is a new algorithm
for indexing diffraction patterns, applicable when the unit cell
geometry is known. Originally based on the PyTorch framework a
dedicated version in CUDA has been developed. The algorithm is
capable of indexing frames in less than a millisecond. It
therefore allows real-time data processing. Not only being fast
its quality is also on par with the XGandalf indexing algorithm. A
user-friendly interface allows the configuration to the particular
needs of an experiment and the inclusion into the CrystFEL data
processing pipeline.
- Gasparotto P, Barba L, Stadler H-C, Assmann G, Mendonça H,
Ashton A, et al. TORO Indexer: A PyTorch-Based Indexing Algorithm
for kHz Serial Crystallography. ChemRxiv. 2023;
doi:10.26434/chemrxiv-2023-wnm9n (submitted to Journal of Applied
Crystallography, in proof).
| Abstract publication | I agree that the abstract will be published on the web site |
|---|
