Richard Zou 73a661abf1 Stop using excess memory in generate_opcheck_tests, re-enable fbgemm TBE tests (#114641) ... Summary: 1. We stop using excess memory in generate_opcheck_tests. This is safe because all the individual test utils already ensure that they do not modify the inputs. 2. We re-enable the fbgemm TBE tests (see internal diff, but all of this is open source). They were previously removed because they OOM'ed when run serially; (1) and (3) cut down the memory usage to ~20gb peak. 3. I needed to skip some newly failing generated tests and also some that had an impact on the memory usage. Test Plan: - run tests Reviewed By: sryap Differential Revision: D51601964 Pull Request resolved: https://github.com/pytorch/pytorch/pull/114641 Approved by: https://github.com/williamwen42		2023-11-29 02:21:13 +00:00
..
__init__.py	[optests] Add dontGenerateOpCheckTests and is_inside_opcheck_mode (#110951)	2023-10-10 21:43:43 +00:00
aot_autograd.py	Use pytree.arg_tree_leaves everywhere (#112394)	2023-10-31 15:57:06 +00:00
autograd_registration.py	Use pytree.arg_tree_leaves everywhere (#112394)	2023-10-31 15:57:06 +00:00
fake_tensor.py	optests improvements based on torchvision usage on nms (#108929)	2023-09-13 13:26:15 +00:00
generate_tests.py	Stop using excess memory in generate_opcheck_tests, re-enable fbgemm TBE tests (#114641)	2023-11-29 02:21:13 +00:00
make_fx.py	operator_compile_check v0 (#103198)	2023-06-14 14:00:14 +00:00