lucidrains / performer-pytorch

An implementation of Performer, a linear attention-based transformer, in Pytorch
MIT License
1.07k stars 143 forks source link

Names `to_k`, `to_q`, `to_v`, `to_out` cause issues #71

Open JamesDeAntonis opened 3 years ago

JamesDeAntonis commented 3 years ago

Hi Lucid,

Hope all is well. Is there any need to have the q, k, v, o be named to_k, to_q, to_v, to_out? This causes lots of incompatibility with huggingface, since they call them q, k, v, o. Any chance we could change this to the HF convention?

Thanks!