1.Have you used Rmse/MAE/HSS and the like on the test set quantitatively to calculate different thresholds?Is it also an improvement over the baseline model? Although the image is indeed clearer from the results of your experiments, are there any cases where the radar echoes are not located correctly?
2.If possible, could you please provide some core code for reference? If so, thank you very much. I try to use the open source PG-Gan, but the effect is really poor, which makes me confused
1.Have you used Rmse/MAE/HSS and the like on the test set quantitatively to calculate different thresholds?Is it also an improvement over the baseline model? Although the image is indeed clearer from the results of your experiments, are there any cases where the radar echoes are not located correctly? 2.If possible, could you please provide some core code for reference? If so, thank you very much. I try to use the open source PG-Gan, but the effect is really poor, which makes me confused