Tommy van der Vorst

> Hi @pixelspark, I've encountered the same error trying to run YOLOv8 via wonnx. Have you had a chance to look into this issue yet? > > If you don't...

While I think we encourage anyone interested to work on this (and I'd be happy to review PRs), this is not currently in scope unfortunately as it is quite a...

I am not sure from the above if wonnx throws an error when just the import is present? If so we can actually make wonnx a bit more lenient and...

@philpax you should be able to try this out now (see limitations above). Looking for some feedback!

> What will the developer experience be like? Is some sort of "interactive" mode planned where i can see the results an operation will have on a tensor while stepping...

> > Our execution happens on the GPU and we generate a single set of commands to execute all ops in one go, so breakpoints are difficult to support. Also...

> Hey i finally got some time to take a look at these changes and i will start to play around a bit. Is there somewhere a list of all...

> I converted a GPT-2 model to ONNX and visualized the operations [here](https://github.com/webonnx/wonnx/assets/65088241/676c145f-5c21-48d3-ab19-aec523fb628c). The main thing still missing are the different shape manipulating operations like "Squeeze, Unsqueez, Concat, Gather" and...

> But i now know what you were referring to with "Replace "op name + attributes" with an actual enum containing all (supported) ops and their parameters". 😅 Yeah, that's...

> Is there a way to instantiate a tensor from a memory address, an offset and dimensions? Or do i have to first load the ggml tensor into lists and...