Pian Pawakapan 1132b6764a [draft export] generate fake outputs when real tensor prop finds mismatches (#139766) ... Currently real tensor tracing raises MetadataMismatchErrors if registered fake kernels don't match the real kernels (e.g. shape, aliasing, dtype, etc.). This adds an option to use fake kernel inference to bypass mismatches - this option defaults to False for real tensor tracing, but is on for draft export. Pull Request resolved: https://github.com/pytorch/pytorch/pull/139766 Approved by: https://github.com/angelayi, https://github.com/zou3519		2024-11-21 08:01:09 +00:00
..
__init__.py
_fake_tensor_utils.py
fake_impls.py	[BE]: Apply PERF401 autofixes from ruff (#140980)	2024-11-20 17:52:07 +00:00
fake_tensor.py	[draft export] generate fake outputs when real tensor prop finds mismatches (#139766)	2024-11-21 08:01:09 +00:00
fake_utils.py	check fake/real mismatches during real tensor prop (#137747)	2024-11-04 23:39:48 +00:00
functional_tensor.py
meta_utils.py	type annotations for meta_utils (#140203)	2024-11-13 20:07:47 +00:00
schema_check_mode.py