Hello, thanks for checking this issue. I use the uploaded code and run UDA on the Office-Home dataset. However, I cannot reproduce the claimed results in the paper. Some of the comparisons are shown below:
Do we need different hyper-params on the Office-Home? Would you mind sharing the pre-trained model or more information about the data split on the Office-Home? Or how can we do to reproduce the results on the Office-Home? Thank you~
Hello, thanks for checking this issue. I use the uploaded code and run UDA on the Office-Home dataset. However, I cannot reproduce the claimed results in the paper. Some of the comparisons are shown below: Do we need different hyper-params on the Office-Home? Would you mind sharing the pre-trained model or more information about the data split on the Office-Home? Or how can we do to reproduce the results on the Office-Home? Thank you~