Matthijs Hollemans
Matthijs Hollemans
It works on the device now with beta 2, but I'm not sure yet if that's because the model now runs on CPU instead of the GPU, or that Metal...
@XBeg9 I haven't had much luck with Instruments and compute shaders. But with GPU Frame Capture in Xcode it should be possible to check. I just haven't had the time...
What are the min and max values for the predicted logits? `atol=3` seems large but if the logits themselves are huge, then a difference of 3 may not be very...
I didn't really look at the problem too closely but if this model runs on the Neural Engine / M1 GPU, then precision is limited to 16-bit floats. It's possible...
I don't think you can use this repo to convert YOLOv3 models.
I don't know, but sometimes the conversion to Core ML does not fill in those output shapes correctly. What is the shape of these outputs when you actually run the...
That is not what I asked for (it's the same information as shown in Xcode). Instead, I'd need to see the shape of the output when you actually run the...
That's what I thought: the shapes are correct. It's just that the converter didn't fill them in right in the mlmodel file. You should be able to use the model...
Ah yes, that makes sense. You can fix this by adding the exact output shape to the mlmodel file (you can read how to do this in my ebook [Core...
Your YOLO model probably doesn't have a 13x13 output grid, or it has fewer detectors per grid, or it doesn't have the same number of classes. You'll need to tweak...