You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
The linear layers defined in the PSP aim to perform a linear transformation to the audio or visual feature, we just opt to implement it by a matrix multiplication without adding bias, you may try to set the bias to True and I guess the result is competitive to (or may be lower than) current setup.
I am afraid that different initialization methods would have some influence on the model training. Obviously, a good initialization method would help to achieve the optimal situation.
Hi,
Thanks for your code and congrats on the paper accept. I was going through your code and your model definition. I have 2 enquiries:
Thanks!
The text was updated successfully, but these errors were encountered: