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

BN layer and dropout #79

Open

zy23456 opened this issue Jun 6, 2020 · 0 comments

zy23456 commented Jun 6, 2020 •

edited

Loading

Thank you very much for your repo, it helped me a lot, but there are only two small doubts

In the backpropagation implementation of the BN layer, is the derivative of Xmean missing one item, because self.stddev_inv also contains mean
In the implementation of dropout, is the meaning of p different from p in the original dropout? The p in the original dropout is the proportion of the inactivated unit. It seems that it is not here, and the signal strength after the dropout is Isn’t it the same as before?

zy23456 changed the title ~~Back propagation at the BN layer~~ BN layer and dropout

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