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

Hard Negative Mining problem #121

Open

shinji-ohkubo opened this issue Nov 18, 2020 · 0 comments

shinji-ohkubo commented Nov 18, 2020

multibox_loss.py
-->num_neg = torch.clamp(self.negpos_ratio * num_pos, max=pos.size(1) - 1)

If positive is 0, negative is also 0, which hinders negative learning and causes over-detection. I think it is correct to set a fixed minimum value.

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