Tommy van der Vorst
Tommy van der Vorst
@LLukas22 still seeing the following:  SHA-256 Hash of the file as I see it (PowerShell `Get-FileHash`): D0E1BC0DEA48252CCE95552DBCA4E74DE9D49024C4583DEDD497359A89B2F9A2 As for MPT:  Do I need to use other files?
@LLukas22 sorry I am stupid - forgot to do a `git submodule update`. Now it all seems to work! Still getting some weird results though:  MPT is fine apparently...
@antimora great to hear! If you haven't seen it yet, you might want to have a look at https://github.com/webonnx/wonnx/pull/170. It is work in progress to offer a non-ONNX builder API...
Hi @AmineDiro! In WONNX, shape inference and model inference are two completely separate components. Model inference requires all shapes to be inferred (which you can do using the shape inference...
Hm, these are always tough issues to figure out... First of all, you should establish that this is not actually some rounding error (differences are expected given wgpu is a...
@maxwellflitton the error indicates an output from an earlier node cannot be found in the ONNX file. Can you share your ONNX file?
We are not very strict at implementing a specific version of the opset currently (in cases where there are differences, the code does sometimes check and behave differently for different...
You might try upgrading wgpu - maybe that fixes it. We are using an old version (can't update just yet because of an issue with wgpu on the web -...
Could you test again with v0.5.0?
The error: ```` Caused by: In Device::create_compute_pipeline Internal error: FXC D3DCompile error (0x80004005): (81,21-35): warning X3550: array reference cannot be used as an l-value; not natively addressable, forcing loop to...