Ramyc

Did you solve it? I meet similar problem.

Sorry but I've been out of this field for a long time and I didn't preserve the weights.

I forget how many pictures I used but I got the best result with 150 epochs. The loss should converge stably. Maybe you can try to use less pictures and...

Sorry for late reply. If you are still looking for SAR and optical pair images dataset,this is dataset I used: https://mediatum.ub.tum.de/1436631