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

Fix Checkpoint issue when using Horovod distributed backend (PyTorchLightning#6947) #6958

Conversation

liob
Copy link
Contributor

@liob liob commented Apr 11, 2021

What does this PR do?

Fixes #6947: ModelCheckpoint failed when using the save_top_k feature because the ReduceOp was not considered when evaluating reduce_sum.

cc @awaelchli

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 update the CHANGELOG? (not for typos, docs, test updates, or internal minor changes/refactorings)

PR review

Anyone in the community is free to review the PR once the tests have passed.
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

…ightning#6947)

Co-Authored-By: Adrian Wälchli <aedu.waelchli@gmail.com>
@codecov
Copy link

codecov bot commented Apr 11, 2021

Codecov Report

Merging #6958 (f10b2a4) into master (6c01608) will increase coverage by 44%.
The diff coverage is 0%.

❗ Current head f10b2a4 differs from pull request most recent head 7d32ceb. Consider uploading reports for the commit 7d32ceb to get more accurate results

@@           Coverage Diff            @@
##           master   #6958     +/-   ##
========================================
+ Coverage      43%     87%    +44%     
========================================
  Files         194     194             
  Lines       12365   12346     -19     
========================================
+ Hits         5330   10781   +5451     
+ Misses       7035    1565   -5470     

@awaelchli awaelchli added checkpointing Related to checkpointing distributed Generic distributed-related topic bug Something isn't working labels Apr 11, 2021
@liob liob marked this pull request as ready for review April 11, 2021 14:53
@awaelchli awaelchli added the ready PRs ready to be merged label Apr 12, 2021
@SeanNaren SeanNaren enabled auto-merge (squash) April 12, 2021 11:05
@SeanNaren SeanNaren merged commit b37b58a into Lightning-AI:master Apr 13, 2021
SeanNaren pushed a commit that referenced this pull request Apr 13, 2021
…ightning#6947) (#6958)

Co-Authored-By: Adrian Wälchli <aedu.waelchli@gmail.com>

Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>

(cherry picked from commit b37b58a)
@SeanNaren SeanNaren mentioned this pull request Apr 13, 2021
awaelchli pushed a commit that referenced this pull request Apr 13, 2021
…ightning#6947) (#6958)

Co-Authored-By: Adrian Wälchli <aedu.waelchli@gmail.com>

Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>

(cherry picked from commit b37b58a)
lexierule pushed a commit that referenced this pull request Apr 14, 2021
…ightning#6947) (#6958)

Co-Authored-By: Adrian Wälchli <aedu.waelchli@gmail.com>

Co-authored-by: Adrian Wälchli <aedu.waelchli@gmail.com>

(cherry picked from commit b37b58a)
@awaelchli awaelchli mentioned this pull request Apr 18, 2021
11 tasks
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
bug Something isn't working checkpointing Related to checkpointing distributed Generic distributed-related topic ready PRs ready to be merged
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Checkpoint issue when using Horovod distributed backend
4 participants