🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/165433

• 📄 Preview Python docs built from this PR
• 📄 Preview C++ docs built from this PR
• ❓ Need help or want to give feedback on the CI? Visit the bot commands wiki

Note: Links to docs will display an error until the docs builds have been completed.

✅ No Failures

As of commit abe388d with merge base fee1ac9 ():
💚 Looks good so far! There are no failures yet. 💚

This comment was automatically generated by Dr. CI and updates every 15 minutes.

@pytorchbot label "topic: not user facing"

cc @fffrog @albanD , PTAL ,thanks.

@zhudada0120 some code conflicts here

I’ve resolved the merge conflicts; ready for review. @fffrog

@pytorchbot merge

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

@pytorchbot revert -m "I think this broke some quantization tests" -c nosignal

The quantization tests only run on a weekly workflow now, I'll add the label to trigger it in OSS
But here's some logs from internal which I assume are going to be the same

FAIL: test_fake_quant_control (test.quantization.core.test_workflow_ops.TestFakeQuantizeOps)
----------------------------------------------------------------------
Traceback (most recent call last):
  File ".../test/quantization/core/test_workflow_ops.py", line 622, in test_fake_quant_control
    _LearnableFakeQuantize.with_args(observer=MovingAverageMinMaxObserver, quant_min=0,
  File "...torch/ao/quantization/observer.py", line 88, in __call__
    return self.p(*args, **keywords)
           ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "...torch/ao/quantization/_learnable_fake_quantize.py", line 67, in __init__
    raise AssertionError("quant_min out of bound")
AssertionError: quant_min out of bound

======================================================================
FAIL: test_fq_serializable_per_tensor (test.quantization.core.test_workflow_ops.TestFakeQuantizeOps)
----------------------------------------------------------------------
Traceback (most recent call last):
  File ".../test/quantization/core/test_workflow_ops.py", line 602, in test_fq_serializable_per_tensor
    fq_module = FakeQuantizeClass(observer, quant_min, quant_max)
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "...torch/ao/quantization/_learnable_fake_quantize.py", line 67, in __init__
    raise AssertionError("quant_min out of bound")
AssertionError: quant_min out of bound

@pytorchbot successfully started a revert job. Check the current status here.
Questions? Feedback? Please reach out to the PyTorch DevX Team

@zhudada0120 your PR has been successfully reverted.

@pytorchbot revert -m "I think this broke some quantization tests" -c nosignal

The quantization tests only run on a weekly workflow now, I'll add the label to trigger it in OSS But here's some logs from internal which I assume are going to be the same

FAIL: test_fake_quant_control (test.quantization.core.test_workflow_ops.TestFakeQuantizeOps)
----------------------------------------------------------------------
Traceback (most recent call last):
  File ".../test/quantization/core/test_workflow_ops.py", line 622, in test_fake_quant_control
    _LearnableFakeQuantize.with_args(observer=MovingAverageMinMaxObserver, quant_min=0,
  File "...torch/ao/quantization/observer.py", line 88, in __call__
    return self.p(*args, **keywords)
           ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "...torch/ao/quantization/_learnable_fake_quantize.py", line 67, in __init__
    raise AssertionError("quant_min out of bound")
AssertionError: quant_min out of bound

======================================================================
FAIL: test_fq_serializable_per_tensor (test.quantization.core.test_workflow_ops.TestFakeQuantizeOps)
----------------------------------------------------------------------
Traceback (most recent call last):
  File ".../test/quantization/core/test_workflow_ops.py", line 602, in test_fq_serializable_per_tensor
    fq_module = FakeQuantizeClass(observer, quant_min, quant_max)
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "...torch/ao/quantization/_learnable_fake_quantize.py", line 67, in __init__
    raise AssertionError("quant_min out of bound")
AssertionError: quant_min out of bound

@clee2000 Sorry for the trouble! The issue has been fixed. Let me know if there are any further issues!

Hi @albanD , sorry about this! Due to a mistake in my Git operations, some irrelevant commits got included in the PR. I’ve already fixed the issue and will be extra careful moving forward. Could you please help clean up the automatically added reviewers and labels? Sorry again for the trouble!

Have checked all the changes, which seems good to me.

Wait for CI passed first

cc @fffrog , I have resolved the code conflicts and need to re-trigger the CI. Thanks

@pytorchbot merge

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Merge failed

Reason: 1 mandatory check(s) failed. The first few are:

• Lint / Link checks / lint-xrefs / lint-xrefs

Dig deeper by viewing the failures on hud

@pytorchbot merge

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team

Merge failed

Reason: 46 mandatory check(s) failed. The first few are:

• pull / linux-jammy-py3.10-clang12 / build
• pull / linux-jammy-cuda12.8-cudnn9-py3.10-clang12 / build
• pull / cuda12.8-py3.10-gcc9-sm75 / build
• pull / linux-jammy-xpu-n-py3.10 / build
• pull / linux-docs / build-docs-cpp-false

Dig deeper by viewing the failures on hud

@pytorchbot rebase

@pytorchbot started a rebase job onto refs/remotes/origin/viable/strict. Check the current status here

Successfully rebased clean-asserts-quantization-other onto refs/remotes/origin/viable/strict, please pull locally before adding more changes (for example, via git checkout clean-asserts-quantization-other && git pull --rebase)

Ho looks like there are some merge conflicts now.

@pytorchbot rebase -b main

@pytorchbot started a rebase job onto refs/remotes/origin/main. Check the current status here

Rebase failed due to Command git -C /home/runner/work/pytorch/pytorch rebase refs/remotes/origin/main pull/165433/head returned non-zero exit code 1

Rebasing (1/2)
Auto-merging torch/ao/quantization/observer.py
Auto-merging torch/ao/quantization/quantizer/x86_inductor_quantizer.py
CONFLICT (content): Merge conflict in torch/ao/quantization/quantizer/x86_inductor_quantizer.py
error: could not apply 21fdb3433f8... [Code Clean] Clean asserts in torch/ao/quantization/*.py and torch/ao/quantization/quantizer/* and torch/ao/quantization/backend_config/*
hint: Resolve all conflicts manually, mark them as resolved with
hint: "git add/rm <conflicted_files>", then run "git rebase --continue".
hint: You can instead skip this commit: run "git rebase --skip".
hint: To abort and get back to the state before "git rebase", run "git rebase --abort".
hint: Disable this message with "git config set advice.mergeConflict false"
Could not apply 21fdb3433f8... # [Code Clean] Clean asserts in torch/ao/quantization/*.py and torch/ao/quantization/quantizer/* and torch/ao/quantization/backend_config/*

Raised by https://github.com/pytorch/pytorch/actions/runs/18961883926

There are still conflicts with main.

Hi, @zhudada0120. You can run git pull upstream main --rebase to solve conflicts locally.

The conflict has been resolved.

Hi, @cyyever @albanD, I think this PR is ready to be merged. Please take a look. Thanks.

@pytorchbot merge

Merge started

Your change will be merged once all checks pass (ETA 0-4 Hours).

Learn more about merging in the wiki.

Questions? Feedback? Please reach out to the PyTorch DevX Team