Skip to content
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.

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

ECE Computation #4

Open
GloryyrolG opened this issue Dec 7, 2022 · 1 comment
Open

ECE Computation #4

GloryyrolG opened this issue Dec 7, 2022 · 1 comment

Comments

@GloryyrolG
Copy link

GloryyrolG commented Dec 7, 2022

Hi Sinz Lab,

Appreciate your code release for the inspiring work.

May I ask a question on the computation of ECE, in

https://github.com/sinzlab/cGNF/blob/b126121a40f96e8086a98c1a4f19604b1071db6f/notebooks/toy_problem/minMPJPE_and_ECE_landscapes.ipynb

Why is it

x_cdf_vals = np.abs(x_cdf_vals - 0.5) * 2

?

Thanks & Regards,

@GloryyrolG
Copy link
Author

Um, I think you implemented an interval prop_type instead of a quantile one.

        prop_type: "interval" to measure observed proportions for centered prediction intervals,
                   and "quantile" for observed proportions below a predicted quantile.

Ref: https://github.com/uncertainty-toolbox/uncertainty-toolbox/blob/f7ca1ec948b71e5214577015b3f831a39785d707/uncertainty_toolbox/metrics_calibration.py#L98

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant