Summary:
`torch.nn` modules in Python save their kwarg options directly as module object attributes, while `torch::nn` modules in C++ save their options inside the `options` field of the module object. This PR tries to map between these two (by using the newly added `options_args` list to discover options arguments in Python module), to make sure options equivalence is properly checked.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/25784
Differential Revision: D17238609
Pulled By: yf225
fbshipit-source-id: 2febd277ddcbe3ab458ac3feaaf93e4c94bb5b98
Summary:
This PR adds Python/C++ API parity tracker at `test/cpp_api_parity/parity-tracker.md`, which currently shows parity status for `torch.nn` modules.
A good amount of line changes here is moving `new_criterion_tests` from `test_nn.py` to `common_nn.py`, so that it can be used in `test_cpp_api_parity.py`.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/25289
Differential Revision: D17188085
Pulled By: yf225
fbshipit-source-id: 33d12fb1a4de2d9147ed09380973f361a3981fdf
Summary:
This PR adds test harness for checking Python / C++ API parity for `torch.nn.Module` subclasses. Under the hood, we use JIT tracing to transfer `nn.Module` state from Python to C++, so that we can test initialization / forward / backward on Python / C++ modules with the same parameters and buffers.
Pull Request resolved: https://github.com/pytorch/pytorch/pull/23852
Differential Revision: D16830204
Pulled By: yf225
fbshipit-source-id: 9b5298c0e8cd30e341a9f026e6f05604a82d6002
