caffe2 | Caffe2 is a lightweight , modular , and scalable deep | Machine Learning library

AVX, AVX2, and FMA are CPU instruction sets and are not related to multi-threading. If the pip package for pytorch/caffe2 used these instructions on a CPU that didn't support them, the software wouldnt work. Pytorch, installed via pip comes with multi-threading enabled though. You can confirm this with torch.__config__.parallel_info()

As the user metatoaster suggested, I did the following:

1. commented this line that I added to ~/.bashrc and sourced it:

   #export PYTHONPATH="~/venv/ipnet/lib/python3.8/site-packages"

2. removed the --user flag here:

   python setup.py install

Torch wheel contains caffe2 directory.

The issue is resolved. I was using the command-line utility suggested on their README. However, it points to their tutorial in deprecated version of the code.

The command-line utility (installed using the pip install onnx-caffe2) still has the _known_opset_version = 3. This was causing the error. After I used the conversion utility through Python APIs in PyTorch library by importing,

An Internal compiler error is always a bug with the compiler. In this case, it's prevented building of a library that is needed later in the build process.

Your options are limited. I suggest trying a different version of Visual Studio.

You should also report this to Microsoft.

I have found a link overcome the issue you are facing. Please read it through it should solve your solve. https://github.com/pytorch/pytorch/issues/14727

Core ML can use the Apple Neural Engine (ANE), which is much faster than running the model on the CPU or GPU. If a device has no ANE, Core ML can automatically fall back to the GPU or CPU.

I haven't really looked into PyTorch Mobile in detail, but I think it currently only runs on the CPU, not on the GPU. And it definitely won't run on the ANE because only Core ML can do that.

Converting models can be a hassle, especially from PyTorch which requires going through ONNX first. But you do end up with a much faster way to run those models.

You could package your application using pyinstaller. This tool packages your app with Python and dependencies and use only the parts you need (simplifying, in reality it's hard to trace your package exactly so some other stuff would be bundled as well).

Also you might be in for some quirks and workarounds to make it work with pytorch and numpy as those dependencies are quite heavy (especially pytorch).

numpy and pytorch are pretty similar feature-wise (as PyTorch tries to be compatible with it) hence maybe you could only use only of them which would simplify the whole thing further

Depending on other parts of your app you may write it (at least neural network) in C++ using PyTorch's C++ frontend which is stable since 1.5.0 release.

Going this route would allow you to compile PyTorch's .cpp source code statically (so all dependencies are linked) which allows you for relatively small binary size (30Mb when compared to PyTorch's 1GB+), but requires a lot of work.

The problem is that the Caffe2 ONNX backend does not yet support the export of the Resize operator.

Please raise an issue on the Caffe2 / PyTorch github -- there's an active community of developers who should be able to address this use case.