torch::deploy (multipy for non-torch uses) is a system that lets you get around the GIL problem by running multiple Python interpreters in a single C++ process.
In this test of autograd, why is result = w.grad commented while keeping another hardcoding result? I have tried to uncomment the former and comment the latter. Then AutoGrad test failed. Is there an issue there?
There is a line of code that seems problematic. https://github.com/pytorch/multipy/blob/cef6cac074fc2d72e7b73e88bf39478d859c6386/multipy/runtime/test_deploy.cpp#L461-L462
In this test of autograd, why is
result = w.grad
commented while keeping another hardcodingresult
? I have tried to uncomment the former and comment the latter. Then AutoGrad test failed. Is there an issue there?