Hi, I am trying to adopt another model as base-encoder, but it seems wrong with the type of q, it reminds that I am using tuple other than the original type torch tensor...
I just wondering if the model can be changed...or the problem is my model?
the following is the forward part of my model:
with regards,
I would be really appreciate if you could give me some advice!
Hi, I am trying to adopt another model as base-encoder, but it seems wrong with the type of q, it reminds that I am using tuple other than the original type torch tensor...
I just wondering if the model can be changed...or the problem is my model? the following is the forward part of my model:
with regards, I would be really appreciate if you could give me some advice!