xyy19920105

I have the same problem. I can use matcaffe like caffe.set_mode_gpu(), but I can't use caffe('set_mode_gpu').

I solved it by changing another version of caffe. The version which has problem is not a release version. you guys can download the lastest release version of caffe.

@sachinky I used the caffe version called caffe-rc2, the latest release version.

@sainisanjay Firstly, i'm sorry for @ the wrong id, cause you have the similar id . You can dowonload that release version at https://github.com/BVLC/caffe/releases, the first one.

How to solve this problem.....I have the same question, while I dont think it's a problem any more.

I also think the code of rank loss is not match with the description in the paper. I think loss_p should be 1 + pivot - score

I have the same error

Hi @ruotianluo , Thank you for your code and benchmark, did you test the adaptive attention on your code?? Could you output the adaptive attention's result?? Thank you again.

Thanks for your reply. Do you think that the adaptive attention model is not good enough as a baseline??