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

[bugfix] Resolve logging reduction when using sync_dist + reduce_fx={mean, max} #9142

Merged
merged 36 commits into from Aug 27, 2021

Conversation

tchaton
Copy link
Contributor

@tchaton tchaton commented Aug 26, 2021

What does this PR do?

See title.

Fixes #9100
Fixes #4395

Does your PR introduce any breaking changes? If yes, please list them.

Before submitting

  • Was this discussed/approved via a GitHub issue? (not for typos and docs)
  • Did you read the contributor guideline, Pull Request section?
  • Did you make sure your PR does only one thing, instead of bundling different changes together?
  • Did you make sure to update the documentation with your changes? (if necessary)
  • Did you write any new necessary tests? (not for typos and docs)
  • Did you verify new and existing tests pass locally with your changes?
  • Did you list all the breaking changes introduced by this pull request?
  • Did you update the CHANGELOG? (not for typos, docs, test updates, or internal minor changes/refactorings)

PR review

Anyone in the community is welcome to review the PR.
Before you start reviewing make sure you have read Review guidelines. In short, see the following bullet-list:

  • Is this pull request ready for review? (if not, please submit in draft mode)
  • Check that all items from Before submitting are resolved
  • Make sure the title is self-explanatory and the description concisely explains the PR
  • Add labels and milestones (and optionally projects) to the PR so it can be classified

Did you have fun?

Make sure you had fun coding 🙃

@tchaton tchaton self-assigned this Aug 26, 2021
@tchaton tchaton added the bug Something isn't working label Aug 26, 2021
@tchaton tchaton added this to the v1.4.x milestone Aug 26, 2021
@carmocca carmocca added the logging Related to the `LoggerConnector` and `log()` label Aug 27, 2021
@mergify mergify bot added the ready PRs ready to be merged label Aug 27, 2021
@tchaton tchaton added the priority: 0 High priority task label Aug 27, 2021
@carmocca carmocca disabled auto-merge August 27, 2021 15:01
@carmocca carmocca enabled auto-merge (squash) August 27, 2021 15:01
@mergify mergify bot removed the has conflicts label Aug 27, 2021
@carmocca carmocca merged commit 045c879 into master Aug 27, 2021
@carmocca carmocca deleted the logging branch August 27, 2021 15:40
ethanwharris pushed a commit that referenced this pull request Aug 31, 2021
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
Co-authored-by: Carlos Mocholi <carlossmocholi@gmail.com>
lexierule pushed a commit that referenced this pull request Sep 1, 2021
Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Rohit Gupta <rohitgr1998@gmail.com>
Co-authored-by: Carlos Mocholi <carlossmocholi@gmail.com>
@ohayonguy
Copy link

ohayonguy commented Sep 14, 2021

I am still observing a difference between the logged loss and the loss returned by training_step:
image

Is this expected? nt_xent_loss should be exactly the same as loss.
When logging a loss in training step, should we even pass sync_dist=True? Or do we only care about doing that for validation steps (for accurate model checkpointing)?

This is my training_step:
image

@tchaton
Copy link
Contributor Author

tchaton commented Sep 14, 2021

Hey @ohayonguy,

Lightning is adding a smoothing on the loss by computing a running mean over the logged loss within the progress bar.
Personally, I find it quite confusing.
If you feel the same, please on an issue and ask for it to be removed :)

Best,
T.C

@carmocca
Copy link
Member

If you feel the same, please on an issue and ask for it to be removed :)

You can comment here: #9372

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working logging Related to the `LoggerConnector` and `log()` priority: 0 High priority task ready PRs ready to be merged
Projects
None yet
5 participants