I have been following the mobile (Caffe2 and Core ML) tutorials from the PyTorch v1.0 tutorials section:
At the moment I have been unable to finish either of them because the export does not work.
For the super-resolution one, I have filed a bug on GitHub: https://github.com/pytorch/pytorch/issues/13836, the ONNXLive one, fails with a segmentation fault when converting to coreml.
I have tried both tutorials with a number of nightly builds, but no success so far.
So, at the moment, I’m not sure how to proceed:
- Has anyone been able to run these successfully? If so, with which package/driver/cuda versions?
- Is there a way to get feedback to see if I can fix these myself? Don’t mind fiddling through the code, but wouldn’t want to spend too much time if it’s a simple package version fix.