New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

questions in complex batchnormalization and calculating E mask in the end #31

Open

iris1997jiatong opened this issue Dec 12, 2024 · 0 comments

iris1997jiatong commented Dec 12, 2024 •

edited

Loading

Hello,

Thanks for the nice code!
I encounter NaN after 30 epochs when I train DCCRN and I try to find the problem.
I have two questions.

when calculating whitening matrix

Can you make sure that s, t and inverse_st are always available to calculate (delta>=0, tau+2s >= 0 and s*t != 0)?
when calculating E mask in the end,

Can you make sure that real phase is never equal to zero such that arctan(imag/real) is always available?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment