yshli issues

Repositories
Issues
Comments

Results 2 issues of


                                            yshli

Could someone explain to me the operation of efficient attention？

``` def relu_linear_att(self, qkv: torch.Tensor) -> torch.Tensor: B, _, H, W = list(qkv.size()) if qkv.dtype == torch.float16: qkv = qkv.float() qkv = torch.reshape( qkv, ( B, -1, 3 * self.dim,...

yshli

Could someone explain to me the operation of efficient attention？

Any idea for plotting a PR curve?