The config of the baseline

[momoyo126]

Thank you for your work. Could you provide the baseline training configuration? I train the minkunet 50 epoches without any DG methods and test it in semanticSTF dataset, but miou is only about 15%.