Thanks for the code.
I changed the dataset. But the adaptation process didn't converge( acc=0.5 represents convergence?), and just get 14% domain adaptation accrucy.
I meet the same question as yours. No matter what i do, the acc just fluctuates between 10% and 15%. Domain discriminator and classifier's game out of control in total
Thanks for the code. I changed the dataset. But the adaptation process didn't converge( acc=0.5 represents convergence?), and just get 14% domain adaptation accrucy.
Epoch [457/600] Step [100/200]:d_loss=0.32023 g_loss=3.27941 acc=0.89000
what parameters I need to change? Should I normalize the source and target dataset? Thank you!