jerrymazeyu / fine-grained2019aaai

pytorch implementation of Fine-grained Recognition: Accounting for Subtle Differences between Similar Classes

fine-grained2019AAAI

pytorch implementation of Fine-grained Recognition: Accounting for Subtle Differences between Similar Classes

fine-grained2019aaai's Issues

Project code

您好，在学习了您的文章后，我认为这是一项很不错的工作，同时很希望能复现您文章中的思路，但过程中遇到一些问题，请问能够提供一下完整的代码吗？十分感谢！

没有看到梯度提升损失那一部分的实现？

您好，看到您的文章里边，只对top-k的类进行优化，觉得想法很好，但是似乎在您release的代码里边没有看到，或者是我没有注意到，能否给一些具体实现，谢谢！

diversification_block

        if index_c + 1 == c:
            end_c = c + 1
        else:
            end_c = (index_c + 1) * block_c
        if index_r + 1 == r:
            end_r = r + 1
        else:
            end_r = (index_r + 1) * block_r
        res[index_r * block_r: end_r, index_c * block_c:end_c] = 1

why if index_c + 1==c:
end_c = c+1 rather than (c+1) * block_c ?

project code

Is this the complete code for reproducing the paper? I think the idea is great, can you provide the complete project code?