Revert "Refactor quant_llm to work with affine quantized tensor"

[jerryzh168]

Reverts #696

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/767

• 📄 Preview Python docs built from this PR

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

✅ No Failures

As of commit 2add1b5 with merge base 0fed444 ():
💚 Looks good so far! There are no failures yet. 💚

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

[jerryzh168]

looks like it breaks CI, it has an error that does not appear in the PR CI: https://github.com/pytorch/ao/actions/runs/10588415262/job/29340895949

[gau-nernst]

I think it might have to do with importing C++ extension. Depending on the import order, it might not get imported

ao/torchao/__init__.py

Line 25 in da8c5b9

from . import _C

Maybe that line should be inside ops.py? Or some kind of guards to prevent undefined ops because they are defined in C++?

[jerryzh168]

yeah let me test putting this in ops.py first, that sounds like a reasonable fix, although historically we have been struggling with this import..