Conversation
Signed-off-by: Huy Do <huydhn@gmail.com>
🔗 Helpful Links🧪 See artifacts and rendered test results at hud.pytorch.org/pr/160935
Note: Links to docs will display an error until the docs builds have been completed. ✅ No FailuresAs of commit 1a3a82d with merge base 4e90441 ( This comment was automatically generated by Dr. CI and updates every 15 minutes. |
| with: | ||
| build-environment: linux-jammy-cuda12.8-py3.10-gcc9-sm90 | ||
| dashboard-tag: training-${{ inputs.training }}-inference-${{ inputs.inference }}-default-${{ inputs.default }}-dynamic-${{ inputs.dynamic }}-cudagraphs-${{ inputs.cudagraphs }}-cppwrapper-${{ inputs.cppwrapper }}-aotinductor-${{ inputs.aotinductor }}-maxautotune-${{ inputs.maxautotune }}-freezing_cudagraphs-${{ inputs.freezing_cudagraphs }}-cudagraphs_low_precision-${{ inputs.cudagraphs }} | ||
| dashboard-tag: training-${{ inputs.training || 'true' }}-inference-${{ inputs.inference || 'true' }}-default-${{ inputs.default || 'true' }}-dynamic-${{ inputs.dynamic || 'true' }}-cudagraphs-${{ inputs.cudagraphs || 'true' }}-cppwrapper-${{ inputs.cppwrapper || 'false' }}-aotinductor-${{ inputs.aotinductor || 'false' }}-maxautotune-${{ inputs.maxautotune || 'false' }}-freezing_cudagraphs-${{ inputs.freezing_cudagraphs || 'false' }}-cudagraphs_low_precision-${{ inputs.cudagraphs || 'false' }} |
There was a problem hiding this comment.
When running this workflow in PR, we need some default values for these inputs. These are what I usually use, but let me know if it makes sense to you.
Signed-off-by: Huy Do <huydhn@gmail.com>
|
@pytorchbot merge -f 'CI only tweak, no need to run trunk jobs' |
Merge startedYour change will be merged immediately since you used the force (-f) flag, bypassing any CI checks (ETA: 1-5 minutes). Please use Learn more about merging in the wiki. Questions? Feedback? Please reach out to the PyTorch DevX Team |
I made a mistake in #160935 removing this condition check. This ran the benchmark job twice for schedule jobs, i.e. https://github.com/pytorch/pytorch/actions/runs/17266546494. This was missed during testing because `pull_request` and `workflow_dispatch` were working ok. Pull Request resolved: #161619 Approved by: https://github.com/anijain2305
After pytorch#160635, I can see dependabot creating the PR to bump `transformers` version at pytorch#160807. This a good start, but there are several tweaks we need: 1. Run inductor tests on the PR including one round of perf benchmark, which is always needed. So, we need `ciflow/inductor` label and a `pull_request` trigger for the benchmark 2. Per @anijain2305 feedback, we don't need to update patch version. So, I add a rule to ignore it. Again, we would need to test this out after this lands. Pull Request resolved: pytorch#160935 Approved by: https://github.com/anijain2305
I made a mistake in pytorch#160935 removing this condition check. This ran the benchmark job twice for schedule jobs, i.e. https://github.com/pytorch/pytorch/actions/runs/17266546494. This was missed during testing because `pull_request` and `workflow_dispatch` were working ok. Pull Request resolved: pytorch#161619 Approved by: https://github.com/anijain2305
After #160635, I can see dependabot creating the PR to bump
transformersversion at #160807. This a good start, but there are several tweaks we need:ciflow/inductorlabel and apull_requesttrigger for the benchmark