Open cannonli7 opened 8 months ago
Thanks for your work. Im a little curious about the code of the gan loss in MAE? But i cannot find the code in this repository. could you help me with that
Hello, I also curious about the gan loss in MAE. I found the process is the same when loading two models in demo.py and I also couldn't find definiton of gan loss in the code. Have you solve the problem?
Thanks for your work. Im a little curious about the code of the gan loss in MAE? But i cannot find the code in this repository. could you help me with that
def forward_loss(self, imgs, pred, mask):
"""
imgs: [N, 3, H, W]
pred: [N, L, p*p*3]
mask: [N, L], 0 is keep, 1 is remove,
"""
target = self.patchify(imgs)
if self.norm_pix_loss:
mean = target.mean(dim=-1, keepdim=True)
var = target.var(dim=-1, keepdim=True)
target = (target - mean) / (var + 1.e-6)**.5
loss = (pred - target) ** 2
loss = loss.mean(dim=-1) # [N, L], mean loss per patch
loss = (loss * mask).sum() / mask.sum() # mean loss on removed patches
return loss
如果不是,就是?我也没有找到明文的gan_loss代码,只能暂且这么认为。
Thanks for your work. Im a little curious about the code of the gan loss in MAE? But i cannot find the code in this repository. could you help me with that