Thank you so much for providing us with the code. It's such great work!
However, I'm curious what's the structure of the encoder for domain adaptation during training? From your paper, I think the shape for the output of the encoder is the same as its input as the dynamic parameters. However, I'm not sure whether it is correct. May I have your suggestions? Thank you!
Thank you so much for providing us with the code. It's such great work!
However, I'm curious what's the structure of the encoder for domain adaptation during training? From your paper, I think the shape for the output of the encoder is the same as its input as the dynamic parameters. However, I'm not sure whether it is correct. May I have your suggestions? Thank you!