Isaac Robinson

@SkalskiP it worked in colab which is non windows. @PranavDV-VCTI are all the machines on which it has failed windows?

Oh apologies! :)

Can you run with CUDA_LAUNCH_BLOCKING and post the result?

If you don't specify a resolution it downsizes the images to 384x384 iirc for nano. Either use medium which is higher res or manually set a resolution. Note that setting...

Non square training thread https://github.com/roboflow/rf-detr/issues/275 Num classes initialization is an ongoing issue that seems to be with the print and not actually with the underlying model. Try inference and see...

Not easy due to some internals of the model. Not impossible though. May need to implement anyway for some internal stuff Btw the bicubic should not make it to the...

Again, that layer shouldn't even be in the exported graph. It's only there to resize the positional embedding for different images resolutions, which is fixed at export time. Are you...

We're aware of the issue, same thing that causes degradation in TRT which is resolved by this approach https://github.com/roboflow/rf-detr/issues/176 but I guess that isn't relevant for torchscript Models trained on...

btw how does it work if you use bfloat16?

new model definition should be much more stable in half precision