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

Fixed GMRAE implementation + Improved for DDP #2096

Merged
merged 9 commits into from
Jul 4, 2021

Conversation

KickItLikeShika
Copy link
Contributor

Fixes #2087, #1284

  • Fixed GMRAE implementation to be somehow similar to EpochMetric, all the predictions and targets gathered first then compute the GMRAE on all the gathered values to avoid the misleading results because of computing the mean over a single batch.
  • Updated the integration tests.
  • Made the metric compatible with DDP and added the necessary tests.

Check list:

  • New tests are added (if a new feature is added)
  • New doc strings: description and/or example code are in RST format
  • Documentation is updated (if required)

@github-actions github-actions bot added the module: contrib Contrib module label Jul 3, 2021
@sdesrozis
Copy link
Contributor

@KickItLikeShika Thank you, it looks good to me.

So finally, \bar{A} means the mean of A in the formula.

Copy link
Collaborator

@vfdev-5 vfdev-5 left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Thanks a lot @KickItLikeShika !

@vfdev-5 vfdev-5 merged commit 4d6d220 into pytorch:master Jul 4, 2021
@KickItLikeShika KickItLikeShika deleted the improve-GMRAE-ddp branch July 4, 2021 13:22
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
module: contrib Contrib module
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Improve GMARE metric computation formula
3 participants