Error reading protobuf spec. validator error: Input confidence must have dataType DOUBLE

[glenn-jocher]

Hello, thanks for the great repo. I'm exporting a YOLOv3-tiny model from PyTorch > ONNX > CoreML using the instructions from your article.

The correctly-compiled mlmodel is here: https://storage.googleapis.com/ultralytics/yolov3-tiny-float32.mlmodel

All successful except that my mlmodel outputs Float32, and on attempting to tie its outputs to the NMS inputs I get the following error.

/Users/glennjocher/PycharmProjects/onnx-coreml/venv/lib/python2.7/site-packages/coremltools/models/model.py:109: RuntimeWarning: You will not be able to run predict() on this Core ML model.Underlying exception message was: Error compiling model: "Error reading protobuf spec. validator error: Input confidence must have dataType DOUBLE".
  RuntimeWarning)

Casting the output to torch.float64 in PyTorch produces an error on the CoreML conversion unfortunately (NotImplementedError: Unsupported ONNX ops of type: Cast), so the conversion needs to be done after the existing mlmodel (but before NMS) in CoreML somehow. What's the easiest way to do this? Thanks! (BTW, I did all the box decoding in PyTorch natively, its only NMS I need to pipeline).

[hollance]

In the mlmodel spec file, simply change the datatype of that multiarray to DOUBLE. I don’t have a code snippet handy, but it’s explained in my book Core ML Survival Guide.

[glenn-jocher]

Ah thank you! I found this code modifies the data types successfully, and the error went away. Yes I might have a go at your book!

spec = coreml_model.get_spec()
for i in range(2):
    spec.description.output[i].type.multiArrayType.dataType = \
        coremltools.proto.FeatureTypes_pb2.ArrayFeatureType.ArrayDataType.Value('DOUBLE')

I noticed my array shapes were off compared to your format as well, so I added a squeeze op (to remove the first dimension) and a transpose op (both in PyTorch) to align to your example. Now my mlmodel looks healthier:

[hollance]

Glad to hear you managed to fix it. You don't actually need to use the squeeze, you can simply drop the first dimension in the spec. (But using a squeeze will work fine too, of course.)