Skip to content

Conversation

@jeejeelee
Copy link
Collaborator

The current CI LoRA tests is quite time-consuming, which hampers the development of LoRA-related features. Based on testing on my local single 3090, the three most time-consuming tests are:

case time
test_punica_variation.py 33minutes
test_layer_variation.py 23minutes
test_punica_sizes.py 10minutes

Therefore, I make the following modifications:

  • For test_punica_variation and test_punica_sizes, I reduce the number of test items as much as possible without impacting the test scope.
  • I delete test_layer_variation directly, there are two main reasons:
    • Various layers and ranks are tested across different model-related tests, making this script somewhat redundant.
    • In this script, PEFT initializes LoRA weights by setting lora_b weights to zero, this may prevent it from achieving its intended test effects.

Additionally, I clean up some redundant code.

ping @Yard1 @simon-mo

@github-actions
Copy link

👋 Hi! Thank you for contributing to the vLLM project.
Just a reminder: PRs would not trigger full CI run by default. Instead, it would only run fastcheck CI which consists a small and essential subset of CI tests to quickly catch errors. You can run other CI tests on top of default ones by unblocking the steps in your fast-check build on Buildkite UI.

Once the PR is approved and ready to go, please make sure to run full CI as it is required to merge (or just use auto-merge).

To run full CI, you can do one of these:

  • Comment /ready on the PR
  • Add ready label to the PR
  • Enable auto-merge.

🚀

@jeejeelee
Copy link
Collaborator Author

/ready

@github-actions github-actions bot added the ready ONLY add when PR is ready to merge/full CI is needed label Aug 10, 2024
@jeejeelee
Copy link
Collaborator Author

jeejeelee commented Aug 10, 2024

https://buildkite.com/vllm/ci-aws/builds/6547# can be used as a comparison

@jeejeelee
Copy link
Collaborator Author

@simon-mo Could you please merge this PR?

@simon-mo simon-mo merged commit 9799280 into vllm-project:main Aug 14, 2024
Alvant pushed a commit to compressa-ai/vllm that referenced this pull request Oct 26, 2024
LeiWang1999 pushed a commit to LeiWang1999/vllm-bitblas that referenced this pull request Mar 26, 2025
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

ready ONLY add when PR is ready to merge/full CI is needed

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants