Aditya Goel

Thanks for managing the release @YUNQIUGUO. Given a PyPI [release](https://pypi.org/project/onnxruntime/) has been made for 1.17, is there a plan to tag and release it on GitHub as well? For context,...

We would also appreciate reintroducing the ParallelExecutor.

> Materializing an entire array as opposed to one element is something that should be a common API across libraries, IMHO, Just wanted to point out that it may be...

> I don't run Windows, so it's not something I would work on. I recommend to use WSL. I'm happy to provide a PR that adds win32 builds in the...

It currently bails node adaptation - see https://github.com/Quantco/spox/pull/146.

Thanks for writing this out. I definitely agree that predictable shape inference performance is too important to give up without a _really_ compelling use case. Did you already do the...

I'm going to close this since https://github.com/Quantco/spox/pull/198 has been merged. We went with a conservative approach that gives up shape inference if loop-carried dependencies change shape from one iteration to...

> The right of doing it is to implement the latest onnx specifications ([onnx/onnx#5874](https://github.com/onnx/onnx/pull/5874)) and then to update onnxruntime to support it. I think an [update](https://github.com/microsoft/onnxruntime/pull/21222) to onnxruntime is pending...