Open ThoamsDong opened 5 years ago
The loss function is "Loss_cel + Loss_abv" but in the code of the model, a gradient reversal layer is used to make it "Loss_cel - Loss_abv". I am still trying to fill the gap between my results and the paper. Any advice is welcome.
Thanks for your work! I find that your code of loss function have some problem which only uses "_Loss_cel + Lossabv". However, the paper proposed "_Loss_cel + Lossabv" and "_Loss_cel - Lossabv". And I get very very low accuracy for sample 4 of task "s2m". It was just about 0.005 and even lower. And task "u2m" and task “m2u” have bad results. How should I deal with it? Do you have any advice?