MoCov2_Pytorch_tutorial

MoCov2_Pytorch_tutorial copied to clipboard

loss_fun(): pos = torch.exp(torch.div(torch.bmm(q.view(N, 1, C), k.view(N, C, 1)).view(N, 1), self.args.moco_t))#N*1 neg = torch.sum(torch.exp(torch.div(torch.mm(q.view(N, C), torch.t(self.queue)), self.args.moco_t)), dim=1)#N #denominator is sum over pos and neg denominator = pos + neg...

chenxi52

The result is not ideal

2

Hi Thanks you very much for your work. I see you here is ResNet18, What is the effect if you change to ResNet50?

luohao136486939