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

[RLlib] Pin Gym Everywhere and turn off gpu for recsim tests #23452

Merged
merged 7 commits into from
Mar 24, 2022

Conversation

avnishn
Copy link
Member

@avnishn avnishn commented Mar 24, 2022

Gym wasn't pinned for some PBT tests, and somehow recsim for torch gpu is broken despite #23276 and #22738

Why are these changes needed?

Related issue number

Checks

  • I've run scripts/format.sh to lint the changes in this PR.
  • I've included any doc changes needed for https://docs.ray.io/en/master/.
  • I've made sure the tests are passing. Note that there might be a few flaky tests, see the recent failures at https://flakey-tests.ray.io/
  • Testing Strategy
    • Unit tests
    • Release tests
    • This PR is not tested :(

@avnishn avnishn marked this pull request as draft March 24, 2022 04:13
@avnishn
Copy link
Member Author

avnishn commented Mar 24, 2022

@avnishn
Copy link
Member Author

avnishn commented Mar 24, 2022

@sven1977 sven1977 marked this pull request as ready for review March 24, 2022 08:17
@sven1977 sven1977 merged commit 9040f54 into ray-project:master Mar 24, 2022
@zhe-thoughts zhe-thoughts added the release-blocker P0 Issue that blocks the release label Mar 24, 2022
@jianoaix
Copy link
Contributor

Cherrypicked to 1.12.0rc1.

architkulkarni pushed a commit that referenced this pull request Apr 8, 2022
* [release tests] Pin gym everywhere (#23349)

* [rllib] Pin gym everywhere (#23384)

This PR Pins gym in the app config.yaml's for rllib and tune so that release tests are no longer broken by the new gym version.

* [RLlib] Pin Gym Everywhere and turn off gpu for recsim tests (#23452)

* [ci] Clean up ray-ml requirements (#23325)

In https://github.com/ray-project/ray/blob/ray-1.11.0/docker/ray-ml/Dockerfile, the order of pip install commands currently matters (potentially a lot). It would be good to run one big pip install command to avoid ending up with a broken env.

Co-authored-by: Kai Fricke <[email protected]>

* Fix merge conflict

* Also copy requirements_train.txt

Co-authored-by: Avnish Narayan <[email protected]>
Co-authored-by: ddelange <[email protected]>
Co-authored-by: Amog Kamsetty <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
release-blocker P0 Issue that blocks the release
Projects
None yet
Development

Successfully merging this pull request may close these issues.

5 participants