mirror of
https://github.com/zebrajr/pytorch.git
synced 2025-12-06 12:20:52 +01:00
e7e10e60e0
Summary: This functionality replaces the Scalar-Tensor builtin operators, with builtin functions. Builtin functions are used in place of operators where one operator can be defined using a composition of another. This simplifies later optimization passes by allowing us to have fewer operator. In the future, builtin functions can be used for other purposes. For example, we can define derivative functions as code rather than building graphs. Pull Request resolved: https://github.com/pytorch/pytorch/pull/12141 Reviewed By: ezyang Differential Revision: D10088065 Pulled By: zdevito fbshipit-source-id: a2acb06346e649c4c8a2fe423b420871161c21cf
14 lines
242 B
C++
14 lines
242 B
C++
#pragma once
|
|
|
|
#include "torch/csrc/WindowsTorchApiMacro.h"
|
|
#include "torch/csrc/jit/script/module.h"
|
|
|
|
namespace torch { namespace jit { namespace script {
|
|
|
|
|
|
TORCH_API const std::vector<Method*>& getAllBuiltinFunctionsFor(Symbol name);
|
|
|
|
|
|
|
|
}}}
|