Run tests on GPUs nightly only

[akihironitta]

What does this PR do?

cc @Borda

Before submitting

• Was this discussed/approved via a Github issue? (no need for typos and docs improvements)
• 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?
• [n/a] Did you write any new necessary tests? [not needed for typos/docs]
• [n/a] Did you verify new and existing tests pass locally with your changes?
• [n/a] If you made a notable change (that affects users), did you update the CHANGELOG?

PR review

• Is this pull request ready for review? (if not, please submit in draft mode)

Anyone in the community is free to review the PR once the tests have passed.
If we didn't discuss your PR in Github issues there's a high chance it will not be merged.

Did you have fun?

Make sure you had fun coding 🙃

[codecov]

Codecov Report

Merging #333 (1480783) into master (2796f37) will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master     #333   +/-   ##
=======================================
  Coverage   88.59%   88.59%           
=======================================
  Files          80       80           
  Lines        4061     4061           
=======================================
  Hits         3598     3598           
  Misses        463      463           

Flag	Coverage Δ
unittests	88.59% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 2796f37...1480783. Read the comment docs.

[akihironitta]

The failing test should be addressed in #334.

FAILED tests/examples/test_scripts.py::test_example[predict-translation.py]

[ethanwharris]

We had this at one point before and the GPU tests continued to run on every branch but it also caused the coverage tracking to break as it thought every GPU coverage report was on the master branch. Don't know what was causing that, but this still seems to be running GPU tests on this branch, so I'm not sure it's doing what we want...

@akihironitta @Borda - It looks like we should be able to set trigger: None, pr: None to stop it from running on non-master branches: https://stackoverflow.com/a/66048645

[edgarriba]

@akihironitta @Borda @ethanwharris we have a similar setup in kornia that could help: https://github.com/kornia/kornia/blob/master/.github/workflows/tests_cpu_versions.yml#L3-L9

[akihironitta]

@ethanwharris Thank you for your investigation! I just turned them off with the following lines.

trigger: none
pr: none

[ethanwharris]

Awesome, LGTM 😃

[Borda]

@akihironitta @Borda @ethanwharris we have a similar setup in kornia that could help: https://github.com/kornia/kornia/blob/master/.github/workflows/tests_cpu_versions.yml#L3-L9

yes but you use GHA and this is Azure pipelines so the syntax and behaviour is a bit different...

[Borda]

@ethanwharris just thinking we shall keep triggers for tags so can use the badge in pypi... what do you think?
cc: @akihironitta

[ethanwharris]

@Borda Yeah, that seems reasonable, we should probably run all tests on releases 😃