hongtaoyang / dac-tensorflow Goto Github PK

Hi, I am very grateful for your source code! I try to implement this code with pytorch vision, but it does't work. I wonder if you have tried this? Can I communicate with you about this more specifically？ Thanks a lot!

@HongtaoYang , I am very grateful for your source code! However, I have found that when applying your model to the cifar10 dataset, the ACC indicator is lower than the result in the paper, and the ARI is close to 0. Could you you help me find the reason? Thanks for your kindness!

请问原作者的论文中提到的余弦相似度计算距离，为什么在您的代码中找不到在哪里实现的呢？

@HongtaoYang , I am very grateful for your source code! However, I have found that your implementation is very sensitive to the parameters of the network, such as :

• In the batch_normalization layer, it must set the trainable=False, because when setting "trainable=True", the results will drop a lot.

• In the conv2d layer, it must set the padding to vaild, and the results will also drop when setting the way of padding to "same".

So, I feel very strange and can't understand about this phenomenon, because I think the key of this algorithm is not the design of the network structure, it shouldn't be such sensitive to the parameters of the network, could you explain this phenomenon in detail? Thanks for your kindness!