Matthijs Hollemans

Yes, you need to rotate the images if you’re holding the phone horizontally.

You need to convert this lambda layer into something CoreML understands. I think Core ML supports space-to-depth, so you’d need to insert such a layer manually at that point. See...

It’s in the tfcoreml repo: https://github.com/tf-coreml/tf-coreml/blob/master/examples/custom_layer_examples.ipynb

There is. In fact, the app already shows how to do this! :-)

That means the Core ML file is not a correct protobuf file, or that the contents in the file are not correct. Most likely something is wrong with the weights...

Study the YOLOv5 source code and see what the differences are in the bounding box decoding. 😃

I was under the impression that this was possible these days, but I haven't used MPSNNGraph in ages, so I'm not sure how to do it.

I think Vision pads the bottom / right side of the image when using scaleFit. So you'll need to figure out by how much the image gets padded. If the...

You should probably change this line, ``` let pointer = UnsafeMutablePointer(OpaquePointer(out.dataPointer)) ``` to: ``` let pointer = UnsafeMutablePointer(OpaquePointer(out.dataPointer)) ```

Naturally, if your model arranges the data differently, you'll also have to change how the code reads that data. ;-)