torch._check_is_size is not being recognized by CoreML partitioner

[Ray-Luo]

🐛 Describe the bug

As titled.
torch._check_is_size(pos_x) is not working with CoreML partitioner, the workaround is to use torch._check(pos_x >= 0).

Versions

PyTorch version: 2.6.0
Is debug build: False
CUDA used to build PyTorch: None
ROCM used to build PyTorch: N/A

OS: macOS 15.3.1 (arm64)
GCC version: Could not collect
Clang version: 11.1.0
CMake version: version 3.31.4
Libc version: N/A

Python version: 3.10.0 (default, Mar 3 2022, 03:54:28) [Clang 12.0.0 ] (64-bit runtime)
Python platform: macOS-15.3.1-arm64-arm-64bit
Is CUDA available: False
CUDA runtime version: No CUDA
CUDA_MODULE_LOADING set to: N/A
GPU models and configuration: No CUDA
Nvidia driver version: No CUDA
cuDNN version: No CUDA
HIP runtime version: N/A
MIOpen runtime version: N/A
Is XNNPACK available: True

CPU:
Apple M1 Max

Versions of relevant libraries:
[pip3] executorch==0.5.0a0+6df7779
[pip3] executorchcoreml==0.0.1
[pip3] numpy==2.0.0
[pip3] torch==2.6.0
[pip3] torchao==0.8.0+gitebc43034
[pip3] torchaudio==2.6.0
[pip3] torchsr==1.0.4
[pip3] torchvision==0.21.0
[conda] executorch 0.5.0a0+6df7779 pypi_0 pypi
[conda] executorchcoreml 0.0.1 pypi_0 pypi
[conda] numpy 2.0.0 pypi_0 pypi
[conda] torch 2.6.0 pypi_0 pypi
[conda] torchao 0.8.0+gitebc43034 pypi_0 pypi
[conda] torchaudio 2.6.0 pypi_0 pypi
[conda] torchsr 1.0.4 pypi_0 pypi
[conda] torchvision 0.21.0 pypi_0 pypi

cc @kimishpatel @YifanShenSZ @cymbalrush @metascroy

[metascroy]

@YifanShenSZ can you have a look?

[YifanShenSZ]

I don't think we have a conclusion on such dynamic shape assertions yet 🤦 Currently, Core ML simply ignores those assertions, so you can try add the assertions from your case to it

Long term solution still needs to be developed, since such treatment will get complained by ExecuTorch AOT: Simple removal of assertions leads to unhappy torch.export shape check