Some problems in the code/modeling.py

[fanbowen232218094636-spec]

Hello, I encountered something I didn't understand in the process of reading the code. If you have time, can you help me point it out?
At line 99 of the code/modeling.py, this is：
if posi_emb is not None:
eps=1e-10
batch_size = key_layer.size(0)
patch = key_layer
ad_out, loss_ad = lossZoo.adv_local(patch[:,:,1:], ad_net, is_source)
entropy = - ad_out * torch.log2(ad_out + eps) - (1.0 - ad_out) * torch.log2(1.0 - ad_out + eps)
entropy = torch.cat((torch.ones(batch_size, self.num_attention_heads, 1).to(hidden_states.device).float(), entropy), 2)
trans_ability = entropy if self.vis else None # [B12197]
entropy = entropy.view(batch_size, self.num_attention_heads, 1, -1)
attention_probs = torch.cat((attention_probs[:,:,0,:].unsqueeze(2) * entropy, attention_probs[:,:,1:,:]), 2)
What I don't understand is, is this place a place to fight against discrimination losses? Which part of Vit did you improve?

[fanbowen232218094636-spec]

Hello, I seem to understand the code above. Is it an improvement of the TAM module you defined? Then I have another question, that is, I want to learn how you can only adapt to the confrontation of discriminators on the basis of VIT? Is the first part of your work in code modeling.py Which part of ?

[viyjy]

Hi, thanks for your interest in our paper.
I don't fully understand "only adapt to the confrontation of discriminators on the basis of VIT", can you elaborate it further? Thanks.