Geoscientific Model Development
Geoscientific Model Development
GMD
Articles | Volume 17, issue 23
Articles | Volume 17, issue 23
https://doi.org/10.5194/gmd-17-8873-2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.
https://doi.org/10.5194/gmd-17-8873-2024
© Author(s) 2024. This work is distributed under
the Creative Commons Attribution 4.0 License.
Articles | Volume 17, issue 23
Model experiment description paper
 | 
13 Dec 2024
Model experiment description paper |  | 13 Dec 2024

Architectural insights into and training methodology optimization of Pangu-Weather

Deifilia To, Julian Quinting, Gholam Ali Hoshyaripour, Markus Götz, Achim Streit, and Charlotte Debus

Viewed

Total article views: 6,896 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
4,989 1,563 344 6,896 223 349
  • HTML: 4,989
  • PDF: 1,563
  • XML: 344
  • Total: 6,896
  • BibTeX: 223
  • EndNote: 349
Views and downloads (calculated since 25 Jun 2024)
Cumulative views and downloads (calculated since 25 Jun 2024)

Viewed (geographical distribution)

Total article views: 6,896 (including HTML, PDF, and XML) Thereof 6,837 with geography defined and 59 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 

Cited

Saved (final revised paper)

Latest update: 08 May 2026
Download
Short summary
Pangu-Weather is a breakthrough machine learning model in medium-range weather forecasting that considers 3D atmospheric information. We show that using a simpler 2D framework improves robustness, speeds up training, and reduces computational needs by 20 %–30 %. We introduce a training procedure that varies the importance of atmospheric variables over time to speed up training convergence. Decreasing computational demand increases the accessibility of training and working with the model.
Read more
Share