Skip to content

[1804] feat(training): refactored profiler pr - #2728

Open
florianscheidl wants to merge 75 commits into
ecmwf:developfrom
florianscheidl:fs/refactor-profiler-pr
Open

[1804] feat(training): refactored profiler pr#2728
florianscheidl wants to merge 75 commits into
ecmwf:developfrom
florianscheidl:fs/refactor-profiler-pr

Conversation

@florianscheidl

@florianscheidl florianscheidl commented Aug 4, 2026

Copy link
Copy Markdown
Contributor

Description

Based on #2504, this PR implements a version with fewer changes to trainer.py. Profiling and performance logging are isolated to the ProfilingTrainer, in a separate script. The latter inherits from the Trainer and adds light wrappers for profiling/logging. This keeps the Trainer light while avoiding drift.

Issue Number

Fixes #1804.

Checklist before asking for review

  • I have performed a self-review of my code
  • My changes comply with basic sanity checks:
    • I have fixed formatting issues with ./scripts/actions.sh lint
    • I have run unit tests with ./scripts/actions.sh unit-test
    • I have documented my code and I have updated the docstrings.
    • I have added unit tests, if relevant
  • I have tried my changes with data and code:
    • I have run the integration tests with ./scripts/actions.sh integration-test
    • (bigger changes) I have run a full training and I have written in the comment the run_id(s): launch-slurm.py --time 60
    • (bigger changes and experiments) I have shared a hegdedoc in the github issue with all the configurations and runs for this experiments
  • I have informed and aligned with people impacted by my change:
    • for config changes: the MatterMost channels and/or a design doc
    • for changes of dependencies: the MatterMost software development channel

@github-actions github-actions Bot added infra Issues related to infrastructure performance Work related to performance improvements labels Aug 4, 2026
@florianscheidl
florianscheidl marked this pull request as ready for review August 4, 2026 13:33
@florianscheidl
florianscheidl requested a review from clessig August 4, 2026 13:33
@florianscheidl florianscheidl changed the title feat(profiling): refactored profiler pr [1804] feat(training): refactored profiler pr Aug 4, 2026
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

infra Issues related to infrastructure performance Work related to performance improvements

Projects

Status: No status

Development

Successfully merging this pull request may close these issues.

Implement torch.profiler

3 participants