Deep-Unsupervised-Domain-Adaptation
Deep-Unsupervised-Domain-Adaptation copied to clipboard
agrija9
Pytorch implementation of four neural network based domain adaptation techniques: DeepCORAL, DDC, CDAN and CDAN+E. Evaluated on benchmark dataset Office31.
I trained the model and generate no_adaption files so how can i plot my acc why ?
I have noticed that most of the UDA experiments on Office 31 or visda provide a train_list.txt and val_list.txt during the dataloader creation stage. Doesn't providing the list for both...
Hello! Within the CDAN function, entropy.detach() was not found. Does that mean the entropy will be affected (as well as the classification result and relevant parameters in model) when loss.backward()...
There maybe a wrong with CORAL Loss `loss = torch.norm(torch.mul((source_covariance-target_covariance), (source_covariance-target_covariance)), p="fro")` It should be `loss = torch.norm((source_covariance-target_covariance), p="fro")`
