Yi Zhang
Yi Zhang
I opened an issue https://github.com/pytorch/pytorch/issues/77223 upstream.
Also I think this is only an optimization when the `transpose` is the only user of the original `permute`. Otherwise, we would be replacing one `transpose` with a `listcontruct` and...
Is the 3d case needed in real model? We could delay implementing it until there is a real use case.
A new torch op `aten::add : (Scalar, Scalar) -> (Scalar)` need to be added for purpose of fully testing this. Refer to existing type promotion examples here: https://github.com/llvm/torch-mlir/blob/main/e2e_testing/torchscript/type_promotion.py
Can you show your changes with a PR and paste the error message?