xyu2

> Hey folks, a potential source of `pythonop` is PyTorch autograd. Concretely, if you customize something with `torch.autograd.Function`, then PyTorch autograd may give you some `pythonop` that cannot be converted....

Thanks for confirming the bug. I found there is a workaround in the coremltools source code to support torch.fft.irfft in the Core ML Framework. [coremltools/converters/mil/mil/ops/defs/complex_dialect_ops.py](https://github.com/apple/coremltools/blob/main/coremltools/converters/mil/mil/ops/defs/complex_dialect_ops.py) The function name is complex_irfft...

I see. Thank you for your information!