Create separate float8 tensor subclass

[jainapurva]

Float8Tensor, a new subclass for float8, independent of AQT. It's a BC breaking change, hence will need to proceed with it, by deprecating the current API, and issuing a user warning about the deprecation. The new API will be available to users for use. After a substancial amount of time, once we know user has replaced the deprecated API with the updated one, we can delete it's code

Stack from ghstack (oldest at bottom):

• -> Create separate float8 tensor subclass #1636
• Add boilerplate code #1635

[pytorch-bot]

🔗 Helpful Links

🧪 See artifacts and rendered test results at hud.pytorch.org/pr/pytorch/ao/1636

• 📄 Preview Python docs built from this PR

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

❌ 12 New Failures, 2 Unrelated Failures

As of commit 6af2da8 with merge base abd41e5 ():

NEW FAILURES - The following jobs have failed:

• Build Docs / build_docs (3.11) (gh)
• Code Analysis with Ruff / build (3.9) (gh)
  torchao/dtypes/floatx/float8_layout.py:26:5: F401 [*] torchao.quantization.quant_primitives.MappingType imported but unused
• PR Label Check / Check PR Labels (gh)
  Process completed with exit code 1.
• Run Float8 Tests / test (SM-89, linux.g6.4xlarge.experimental.nvidia.gpu, --pre torch==2.7.0.dev20250122 --index-url... / linux-job (gh)
  RuntimeError: Command docker exec -t 92e68af7b506b82dd582893efe35b14eeff36ee666250b985731369d2135b5b5 /exec failed with exit code 2
• Run Regression Tests / test (CPU 2.3, linux.4xlarge, torch==2.3.0 --index-url https://download.pytorch.org/whl/cpu, cpu) / linux-job (gh)
  RuntimeError: Command docker exec -t a64705491f71fb0f71ea19dbe23faaff578bcda0792982a5f7d2db77e898922b /exec failed with exit code 2
• Run Regression Tests / test (CPU 2.4, linux.4xlarge, torch==2.4.0 --index-url https://download.pytorch.org/whl/cpu, cpu) / linux-job (gh)
  RuntimeError: Command docker exec -t 80c640f351aa3d10e2d4eaa1f1934b82e2c08b82ddbe343d1413eb30277b3074 /exec failed with exit code 2
• Run Regression Tests / test (CPU 2.5.1, linux.4xlarge, torch==2.5.1 --index-url https://download.pytorch.org/whl/cpu, cpu) / linux-job (gh)
  RuntimeError: Command docker exec -t 9d82fb133762905df9a62b13ec08d45239d75e6f9e84e32492651bcd29f9a7e2 /exec failed with exit code 2
• Run Regression Tests / test (CUDA 2.3, linux.g5.12xlarge.nvidia.gpu, torch==2.3.0, cuda, 12.1) / linux-job (gh)
  RuntimeError: Command docker exec -t bcef24a8031a36624534d7e600214f100842e2426a818c5872ffb3dafbde882a /exec failed with exit code 2
• Run Regression Tests / test (CUDA 2.5.1, linux.g5.12xlarge.nvidia.gpu, torch==2.5.1 --index-url https://download.pytorch... / linux-job (gh)
  RuntimeError: Command docker exec -t c5b8fa316ff35cae40f938b0b12fa130e7a1d2d449df8b23422d50637855d28a /exec failed with exit code 2
• Run Regression Tests / test-nightly (CPU Nightly, linux.4xlarge, --pre torch==2.7.0.dev20250122 --index-url https://down... / linux-job (gh)
  RuntimeError: Command docker exec -t ea9a9dc628b1537ccad317a490c91e3bf3e576ddcfc394f636c4f5c767c6a718 /exec failed with exit code 2
• Run Regression Tests / test-nightly (CUDA Nightly, linux.g5.12xlarge.nvidia.gpu, --pre torch==2.7.0.dev20250122 --index-... / linux-job (gh)
  RuntimeError: Command docker exec -t 675aa4c43040ed0cc8e9869b0611123cb36f93ccdf28896a92e7da0ad1d79723 /exec failed with exit code 2
• Run TorchAO Experimental Tests / test (macos-14) (gh)
  AttributeError: type object 'AffineQuantizedTensor' has no attribute 'from_hp_to_floatx'

FLAKY - The following jobs failed but were likely due to flakiness present on trunk:

• Run Float8 Tests / test (H100, linux.aws.h100, --pre torch torchvision torchaudio --index-url https://download.pytor... / linux-job (gh) (detected as infra flaky with no log or failing log classifier)
• Run Regression Tests / test (CUDA 2.4, linux.g5.12xlarge.nvidia.gpu, torch==2.4.0, cuda, 12.1) / linux-job (gh) (detected as infra flaky with no log or failing log classifier)

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

[vkuzo]

Since this is BC-breaking, it would be great to see a more thorough writeup in the PR summary of why we want this and how this fits into the overall strategy of torchao.

[jainapurva]

Since this is BC-breaking, it would be great to see a more thorough writeup in the PR summary of why we want this and how this fits into the overall strategy of torchao.

Yes, this PR is still a work in progress. Will add a thorough write-up