Cannot reproduce the results on office31 and visda (vanilla)

[chester256]

Hi, the paper says you have used 10-crop during evaluation. But it didn't appear in the code. In addition, I cannot reproduce the results on both office31 and visda:

1. For office31, I have tried 10-crop evalution and the result is 86.1. SAFN+ENT is 87.6 reported by the paper.
2. For visda, I have tried both 10-crop and your default eval code (which is not 10-crop), and the results are 69.8 and 71.7 respectively (resnet101 base), which is far from the result reported by the paper.
   All the hyperparameters I have used are the defaults in the code. I have not tried imageCLEF. Hope you can help, thanks.

[chester256]

Sorry that I found the base I have used is resnet50 in visda. I will update this issue if the problem still exist.

[jihanyang]

Hello, due to my recent plans, I don't have time and GPU to reproduce these experiments. However, you can send me an email and I would provide our 10-crop code for you. I hope that would be helpful.