Skip to content

Enable rendered public API docs QA#231

Draft
ChrisRackauckas-Claude wants to merge 1 commit into
SciML:mainfrom
ChrisRackauckas-Claude:agent/rendered-public-docs-qa-v2
Draft

Enable rendered public API docs QA#231
ChrisRackauckas-Claude wants to merge 1 commit into
SciML:mainfrom
ChrisRackauckas-Claude:agent/rendered-public-docs-qa-v2

Conversation

@ChrisRackauckas-Claude

Copy link
Copy Markdown
Member

Ignore this PR until reviewed by @ChrisRackauckas.

Summary

  • enable SciMLTesting rendered public API docs QA
  • keep DEQs as a targeted rendered ignore because it is a module-valued alias already documented in prose on the API page and Documenter canonicalizes its @docs entry to the module binding

Validation

  • git diff --check passed
  • Runic v1.7.0 on test/qa/qa.jl passed
  • Direct QA on Julia +1.10 passed: Quality Assurance | 17 pass, 17 total; doctests passed 1 pass
  • Default Pkg.test() on Julia +1.10 was attempted with timeout 3600; it timed out during Pkg.precompile before reaching test/runtests.jl

No [sources] or path source entries were added.

Co-Authored-By: Chris Rackauckas <accounts@chrisrackauckas.com>
@ChrisRackauckas-Claude

Copy link
Copy Markdown
Member Author

CI note: the GPU lane failure is outside this one-file QA change. The failed job is tests / GPU (julia 1, self-hosted Linux X64 gpu gpu-v100), with the final summary Layers Tests | 734 pass, 36 error, 770 total; the errors are all in the CUDA DEQ convolution-shaped (3, 3, 1, 3) cases across solver/model variants.

The rendered public API docs QA path passes locally on Julia +1.10: Quality Assurance | 17 pass, 17 total, and doctests pass. I did not change tests or backend behavior for this PR.

@ChrisRackauckas-Claude

Copy link
Copy Markdown
Member Author

Current red check is not rendered-docs/QA. The failing job is tests / GPU, with repeated cuDNN convolution execution errors in test/layers_tests.jl for x_size: (3, 3, 1, 3):

CUDNNError: CUDNN_STATUS_EXECUTION_FAILED_CUDART (code 5003)

Summary from the GPU job: DEQ | 540 pass, 36 error, 576 total; Multiscale DEQ | 194 pass, 194 total.

I did not change code here because that is a GPU/cuDNN behavioral failure, not docs/docstrings/rendered public API coverage. Documentation is still in progress on the current poll.

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.

2 participants