Open majing921201 opened 4 weeks ago
Is this issue to track something beyond what's being currently done in https://github.com/intel/torch-xpu-ops/pull/422?
@majing921201 please help to double check it. If this issue have been done, please close it
CUDA add one kernel for performance optimization to support partial channel last case. https://github.com/pytorch/pytorch/blob/df85f34a14dd30f784418624b05bd52b12ab8b0b/aten/src/ATen/native/cuda/UpSampleBilinear2d.cu#L290
performance enhancement is not in 2.5 scope