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

dvc - 10427 Clean Environment Sandbox #1

Open
wants to merge 7 commits into
base: main
Choose a base branch
from

Conversation

nablabits
Copy link
Owner

@nablabits nablabits commented Sep 11, 2024

Related to:

This is just a clean environment where reviewers can check that the approach works as it takes some time to config it from the scratch. You will find more context in this comment but below are the instructions to reproduce and verify the solution

How to Test this

  • Clone the repo
  • Install deps
  • Checkcout the branch. Make sure the submodules dvc & dvc-task are correctly imported

Reproduce the Issue

  • Queue a couple of experiments and run them
dvc exp run --queue --set-param "train.fine_tune_args.epochs=2,3" && \
dvc queue start -j 2 -v
  • Give it some time to complete, you can check this with tail .dvc/tmp/exps/celery/dvc-exp-worker-1.out if something like this appears, it means that they have finished:
[2024-09-05 08:05:31,883: INFO/MainProcess] done
2024-09-05 08:05:31,885 DEBUG: Analytics is enabled.
[2024-09-05 08:05:31,885: DEBUG/MainProcess] Analytics is enabled.
2024-09-05 08:05:31,986 DEBUG: Trying to spawn ['daemon', 'analytics', '/tmp/tmpol8kssna', '-v']
[2024-09-05 08:05:31,986: DEBUG/MainProcess] Trying to spawn ['daemon', 'analytics', '/tmp/tmpol8kssna', '-v']
2024-09-05 08:05:31,999 DEBUG: Spawned ['daemon', 'analytics', '/tmp/tmpol8kssna', '-v'] with pid 594310
[2024-09-05 08:05:31,999: DEBUG/MainProcess] Spawned ['daemon', 'analytics', '/tmp/tmpol8kssna', '-v'] with pid 594310
(example-experiments)
  • Verify that a couple of files with this body remain in in:
grep 'eyJtZXRob2QiOiAic2h1dGRvd24iLCAiYXJndW1lbnRzIjoge30sICJkZXN0aW5hdGlvbiI6IG51bGwsICJwYXR0ZXJuIjogbnVsbCwgIm1hdGNoZXIiOiBudWxsfQ==' \
.dvc/tmp/exps/celery/broker/in/*

Verify the solution

  • Override the default source for dvc-task that will contain the changes:
export PYTHONPATH="`pwd`/sources/dvc-task/src"
  • Clean everything from previous round and start over
dvc exp clean && dvc queue remove --all && \
rm -rf .dvc/cache/runs && \
echo '' > .dvc/tmp/exps/celery/*.out && \
dvc exp run --queue --set-param "train.fine_tune_args.epochs=2,3" && \
dvc queue start -j 2 -v
  • Verify that no file with this body remains in in:
grep 'eyJtZXRob2QiOiAic2h1dGRvd24iLCAiYXJndW1lbnRzIjoge30sICJkZXN0aW5hdGlvbiI6IG51bGwsICJwYXR0ZXJuIjogbnVsbCwgIm1hdGNoZXIiOiBudWxsfQ==' \
.dvc/tmp/exps/celery/broker/in/*

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant