view article
Figure 8
Comparison of prediction of temperature metrics from trained GPR surrogate models using simulated diffraction data input reserved from model training. The metrics are the (a) mean TMean, (b) maximum TMax, (c) minimum TMin and (d) median TMedian of the temperature distribution present in the diffraction volume. The red error bars correspond to the square root of the variance (standard deviation) of the GPR predictions. A linear regression line has been fitted to the testing data and GPR predictions and is shown with a black line. The dashed blue line corresponds to perfect correlation between reserved testing data and GPR model predictions.

Journal logoJOURNAL OF
APPLIED
CRYSTALLOGRAPHY
ISSN: 1600-5767
Volume 56| Part 4| August 2023| Pages 1131-1143
https://doi.org/10.1107/S1600576723005198
Follow J. Appl. Cryst.
Sign up for e-alerts
E-alerts
Follow J. Appl. Cryst. on Twitter
Twitter
Follow us on facebook
Facebook
Sign up for RSS feeds
RSS