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

RGCA: Backlit regularization should go on until end of training #11

Open

una-dinosauria opened this issue Aug 24, 2024 · 0 comments

Contributor

una-dinosauria commented Aug 24, 2024 •

edited

Loading

Currently, backlit reg is configured like this:

    backlit_reg:
      schedule:
        class_name: ca_code.loss.StepWeightSchedule
        start: 500000
        end: 600000
        value: 1.0

If we train for longer (eg, 800k iterations) then backlit regularization turns off. We should enable it until the end of training.

Backlit artifacts:

_point.mp4

The text was updated successfully, but these errors were encountered:

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