Compute AUROC across ranks correctly

facebookresearch / dlrm

An implementation of a deep learning recommendation model (DLRM)

MIT License

3.75k stars 837 forks source link

Closed janekl closed 1 year ago

janekl commented 1 year ago

I'm fixing the issue with computing auroc_result: it was computed at each rank individually instead of the proper aggregate result.

facebook-github-bot commented 1 year ago

@colin2328 has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator.