Closed BowenBao closed 10 months ago
Does it make any difference on torchlib implementation if the op is meant for CUDA? If not, I can take this real quick.
It appears only in cuda export, other than that I don't feel in general there is a difference for torchlib impl.
From models. This op is emitted from cuda export only.
From https://github.com/microsoft/onnx-converters-private/issues/196
cc @justinchuby