frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-11-04 16:04:58 +08:00
Code Issues Packages Projects Releases Wiki Activity

Improve opcheck docs. (#134692)

Browse Source 
Fixes https://github.com/pytorch/pytorch/issues/134119
From user feedback, it's difficult to understand what the tests do. We
clarify the docs more.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/134692
Approved by: https://github.com/albanD
ghstack dependencies: #134466, #134490, #134491, #134690
This commit is contained in:
rzou
2024-08-28 10:17:01 -07:00
committed by PyTorch MergeBot
parent afc76c6f2d
commit dd443f418a
1 changed files with 27 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

36
torch/library.py
View File

@ -1228,17 +1228,35 @@ def opcheck(
``opcheck`` tests these metadata and properties. ``opcheck`` tests these metadata and properties.
Concretely, we test the following: Concretely, we test the following:
- test_schema: if the operator's schema is correct.
- test_autograd_registration: if autograd was registered correctly. - test_schema: If the schema matches the implementation of
the operator. For example: if the schema specifies a Tensor is mutated,
then we check the implementation mutates the Tensor. If the schema
specifies that we return a new Tensor, then we check that the
implementation returns a new Tensor (instead of an existing one or
a view of an existing one).
- test_autograd_registration: If the operator supports training
(autograd): we check that its autograd formula is registered via
torch.library.register_autograd or a manual registration to one
or more DispatchKey::Autograd keys. Any other DispatchKey-based
registrations may lead to undefined behavior.
- test_faketensor: If the operator has a FakeTensor kernel - test_faketensor: If the operator has a FakeTensor kernel
(and if it is correct). The FakeTensor kernel is necessary ( (and if it is correct). The FakeTensor kernel is necessary (
but not sufficient) for the operator to work with PyTorch compilation but not sufficient) for the operator to work with PyTorch compilation
APIs (torch.compile/export/FX). APIs (torch.compile/export/FX). We check that a FakeTensor kernel
(also sometimes known as a meta kernel) was registered for the
operator and that it is correct. This test takes the result of
running the operator on real tensors and the result of running
the operator on FakeTensors and checks that they have the same
Tensor metadata (sizes/strides/dtype/device/etc).
- test_aot_dispatch_dynamic: If the operator has correct behavior - test_aot_dispatch_dynamic: If the operator has correct behavior
with PyTorch compilation APIs (torch.compile/export/FX). with PyTorch compilation APIs (torch.compile/export/FX).
This checks that the outputs (and gradients, if applicable) are the This checks that the outputs (and gradients, if applicable) are the
same under eager-mode PyTorch and torch.compile. same under eager-mode PyTorch and torch.compile.
This test is a superset of ``test_faketensor``. This test is a superset of ``test_faketensor`` and is an e2e test;
other things it tests are that the operator supports
functionalization and that the backward pass (if it exists) also
supports FakeTensor and functionalization.
For best results, please call ``opcheck`` multiple times with a For best results, please call ``opcheck`` multiple times with a
representative set of inputs. If your operator supports representative set of inputs. If your operator supports