您好，想问下提供的resnet或mobilnetv2结构，您有训练测试达到很高的acc吗？还是里面的参数需要自己继续调整。我直接用mbv2训练自己的分类任务，效果不是很好

Hi, thanks for opening the source code. I read the paper, I find you use logits and features before pooling to perform diffusion. but for the logits, I guess the dimension is [B, C] B is the batch size, and C is the class number. This will cause a dimension mismatch in autoencoder, how to solve it. Thanks for your reply.

I want to use Lightvit to write the net structure like this :

Can you tell me how to write the decoder, please?

不知道什么原因，我的ddim_loss,一直在跳，您知道这是为什么吗

Hello,
When you split global token and image token from the input x, shouldn't it be split into [B, :NT, C] and [B, NT: , C]?
But the code in the forward_feature function, it is split from the channel dim for x_glb.

So, assuming x has the shape of [1,3134,64], then global token shape will be [1,8,64] and image token shape will be [1,3136,64].
Please let me know if I am wrong.

Could I use DIST in RetinaFace?

RetinaFace have only 2class(face, not face). so Pearson's correlation coefficient seems to be inefficient.

In summary, if the class is small, the dist is inefficient. Especially in the case of binary, it looks more inefficient.
I wonder if the above opinion is correct.

Hello author, the imagenet dataset I downloaded does not have a meta/train.txt file, can you provide me with a file?

Can I replace the DBB module and model with my own reparameterization module and model?

Hi, sir, the classification lack edgenn model, can you upload it.