Operational Dst index prediction model based on combination of artificial neural network and empirical model | Journal of Space Weather and Space Climate

Topical Issue - Space Weather research in the Digital Age and across the full data lifecycle

Open Access

Table 2

Optimized hyperparameters of ANN models

	Hidden layer node number	Learning rate	Learning cycle
1-hour prediction	13	0.0008	1500
2-hour prediction	50	0.001	1400
3-hour prediction	7	0.01	1500
4-hour prediction	41	0.001	1500
5-hour prediction	30	0.001	1300
6-hour prediction	45	0.001	1200
7-hour prediction	40	0.0001	1400
8-hour prediction	25	0.0001	1200
9-hour prediction	25	0.0001	1200
10-hour prediction	25	0.0001	1200
11-hour prediction	25	0.0001	1200
12-hour prediction	25	0.0001	1200
13-hour prediction	25	0.0001	1200
14-hour prediction	10	0.0001	1300
15-hour prediction	11	0.0001	1300
16-hour prediction	11	0.0001	1300
17-hour prediction	11	0.0001	1300
18-hour prediction	11	0.0001	1300
19-hour prediction	11	0.0001	1300
20-hour prediction	11	0.0001	1300
21-hour prediction	11	0.0001	1300
22-hour prediction	11	0.0001	1300
23-hour prediction	11	0.0001	1300
24-hour prediction	11	0.0001	1300

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.