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

Dropout applied after instead of before the random projection (linear) layer #7

Open

sms821 opened this issue Sep 2, 2021 · 0 comments

sms821 commented Sep 2, 2021 •

edited

Loading

I needed a small clarification.

In the forward() method of class Conv2dDCLLLayer, dropout has been applied after the linear layer

dcll/dcll/pytorch_libdcll.py

Line 540 in 0618dff

pvoutput = self.dropout(self.i2o(flatten))

This is different from the network described in the paper (Table 2 in https://www.frontiersin.org/articles/10.3389/fnins.2020.00424/full) in which dropout is shown before the linear layer.

Is this intentional? Does the location of dropout have a major effect on network accuracy?

Thanks

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