test: use ParallelTestRunner

[avik-pal]

No description provided.

[gemini-code-assist]

Summary of Changes

Hello @avik-pal, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request undertakes a significant refactoring of the LuxCore package's test suite. It transitions from a single, large test file to a modular structure, leveraging ParallelTestRunner to discover and execute tests from individual files. This change aims to enhance the maintainability, organization, and execution speed of the test infrastructure.

Highlights

• Test Suite Refactoring: The existing monolithic runtests.jl file has been broken down into multiple smaller, focused test files, improving organization and maintainability.
• Parallel Test Execution: Integration of ParallelTestRunner enables parallel execution of tests, which is expected to significantly speed up the testing process.
• New Test Files: Dedicated test files have been introduced for AbstractLuxContainerLayer, AbstractLuxLayer, AbstractLuxWrapperLayer interfaces, common test utilities, miscellaneous functionalities, and quality assurance checks.
• Dependency Update: The ParallelTestRunner package has been added as a new dependency in Project.toml to support the new testing infrastructure.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request refactors the LuxCore test suite to use ParallelTestRunner. This is a great improvement that splits a large test file into smaller, more focused files, which should improve maintainability and allow for parallel test execution. During the review, I noticed a couple of pre-existing bugs in the test logic that were carried over into the new files. I've provided suggestions to fix these issues.

[github-actions]

Benchmark Results (Julia v1.12)

Time benchmarks

	main	fa2639d...	main / fa2639d...
basics/MHA	3.3 ± 0.3 μs	3.3 ± 0.3 μs	0.998 ± 0.13
basics/MHA (first run)	3.53 ± 0.38 μs	3.47 ± 0.38 μs	1.02 ± 0.15
basics/MHA reactant	0.0636 ± 0.0066 ms	0.0647 ± 0.005 ms	0.983 ± 0.13
basics/MHA reactant (comp + run)	0.181 ± 0.0058 s	0.181 ± 0.0054 s	1 ± 0.044
basics/conv	23.1 ± 64 μs	21.8 ± 9.5 μs	1.06 ± 3
basics/conv (first run)	24.4 ± 1e+02 μs	22.2 ± 8.6 μs	1.1 ± 4.7
basics/conv reactant	0.0511 ± 0.0025 ms	0.0515 ± 0.0043 ms	0.993 ± 0.097
basics/conv reactant (comp + run)	0.137 ± 0.004 s	0.139 ± 0.0023 s	0.981 ± 0.033
basics/dense	0.135 ± 0.005 μs	0.138 ± 0.004 μs	0.978 ± 0.046
basics/dense (first run)	0.136 ± 0.006 μs	0.146 ± 0.028 μs	0.932 ± 0.18
basics/dense reactant	0.05 ± 0.0024 ms	0.0514 ± 0.0018 ms	0.972 ± 0.058
basics/dense reactant (comp + run)	0.118 ± 0.0039 s	0.119 ± 0.0017 s	0.999 ± 0.036
time_to_load	0.615 ± 0.0064 s	0.624 ± 0.009 s	0.985 ± 0.018

Memory benchmarks

	main	fa2639d...	main / fa2639d...
basics/MHA	0.085 k allocs: 5.75 kB	0.085 k allocs: 5.75 kB	1
basics/MHA (first run)	0.085 k allocs: 5.75 kB	0.085 k allocs: 5.75 kB	1
basics/MHA reactant	19 allocs: 0.578 kB	19 allocs: 0.578 kB	1
basics/MHA reactant (comp + run)	18.5 k allocs: 1.83 MB	18.5 k allocs: 1.83 MB	1
basics/conv	0.039 k allocs: 4.55 kB	0.039 k allocs: 4.55 kB	1
basics/conv (first run)	0.039 k allocs: 4.55 kB	0.039 k allocs: 4.55 kB	1
basics/conv reactant	15 allocs: 0.438 kB	15 allocs: 0.438 kB	1
basics/conv reactant (comp + run)	6.76 k allocs: 1.29 MB	6.76 k allocs: 1.29 MB	1
basics/dense	2 allocs: 0.109 kB	2 allocs: 0.109 kB	1
basics/dense (first run)	2 allocs: 0.109 kB	2 allocs: 0.109 kB	1
basics/dense reactant	15 allocs: 0.422 kB	15 allocs: 0.422 kB	1
basics/dense reactant (comp + run)	6.19 k allocs: 1.25 MB	6.19 k allocs: 1.25 MB	1
time_to_load	0.145 k allocs: 11 kB	0.145 k allocs: 11 kB	1

[avik-pal]

/gemini review

[gemini-code-assist]

Code Review

This pull request refactors the testing infrastructure to use ParallelTestRunner across LuxCore, MLDataDevices, and WeightInitializers. This involves splitting large test files into smaller, more manageable ones, and updating the runtests.jl files to use the new runner. The CI configuration in Buildkite is also updated.

My main feedback is regarding the Buildkite pipeline configuration, where the updated path triggers seem to miss the test directories within the sub-packages, which could prevent CI from running when only tests are modified.

On a positive note, this refactoring also implicitly fixes a bug in the Chain2 test helper within LuxCore's tests.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 75.71%. Comparing base (7900af5) to head (ad72396).

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #1650      +/-   ##
==========================================
- Coverage   76.79%   75.71%   -1.08%     
==========================================
  Files         174      174              
  Lines        7219     7211       -8     
==========================================
- Hits         5544     5460      -84     
- Misses       1675     1751      +76     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.