Skip to content

Fix worker stop loop and results queue handling#2

Open
samblouir wants to merge 5 commits intomainfrom
sam/find-bug-causing-training-to-hang
Open

Fix worker stop loop and results queue handling#2
samblouir wants to merge 5 commits intomainfrom
sam/find-bug-causing-training-to-hang

Conversation

@samblouir
Copy link
Copy Markdown
Owner

Summary

  • ensure broadcast of instructions to each worker
  • send sentinel per worker when controller closes
  • avoid re-enqueueing instructions inside Worker
  • check for stop signal while retrying to push results

Testing

  • python -m unittest discover birdie_rl/tests (fails: ModuleNotFoundError: No module named 'numpy')

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant