view article
Figure 3
Fitting performance of the neural network model on a DIP film grown at 303 K with a deposition rate of 1 Å min−1. (a)–(c) Comparison of the film parameters determined by the neural network with results from LMS fitting with human supervision at different times during growth. The shaded area marks films with thicknesses below 20 Å, where the network has not been trained and consistently yields thick films with high roughness. (d) Overlay of the experimental XRR data with data simulated using the parameters determined by the NN at different times during growth.

Journal logoJOURNAL OF
APPLIED
CRYSTALLOGRAPHY
ISSN: 1600-5767
Volume 52| Part 6| December 2019| Pages 1342-1347
https://doi.org/10.1107/S1600576719013311
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