About the performance on DA datasets like Digits, Office, VisDA-C and etc

[fumyou13]

Hi, thanks for your great work.
I re-implement Tent based on Transfer-Learning-Library to evaluate its performance on DA datasets.
And I only update BN layers as recommented in the paper.
However, I find the accuracy is really low. Could you please release the code to help reproducing the results on Digits and VisDA. Thanks a lot.