Prediction of the SYM-H Index Using a Bayesian Deep Learning Method with Uncertainty Quantification

SYMHnet network version in this repository has the full parameters including m=10 sequence in the modeling. This version requires powerful GPU enabled machine with large memory size. If this is not available, you my use the light-version from https://doi.org/10.5281/zenodo.11211095 which includes m=1 sequnce in the modeling, the light-version is almost the same as the full version with slightly less accuracy.