pytorch

af1bd88f - Allow scalars for aliased binary ops {`multiply`, `subtract`, `divide`} (#65937)

Allow scalars for aliased binary ops {`multiply`, `subtract`, `divide`} (#65937) Summary: https://github.com/pytorch/pytorch/issues/65868 pointed out that the "long-form" versions of some binary ops like `mul`, `sub`, and `div` don't match their alias's behavior when it comes to handling scalar inputs. This PR adds the missing registration in `python_arg_parser.cpp` to resolve this. CC ptrblck ngimel Pull Request resolved: https://github.com/pytorch/pytorch/pull/65937 Reviewed By: malfet Differential Revision: D32156580 Pulled By: ngimel fbshipit-source-id: b143cf7119a8bb51609e1b8734204edb750f0210