question about attention block

[xiaohu2015]

why we use torch.baddbmm to do query @ key?
https://github.com/huggingface/diffusers/blob/main/src/diffusers/models/attention.py#L640

[patrickvonplaten]

In our experiments this was the fastest way of doing the query @ key computation :-) See:
#371
#511