Matthijs Hollemans
Matthijs Hollemans
Indeed you would have to manage all that stuff yourself. Edit: It might be useful if we provided some Swift wrapper code for this that would hide the complexity (since...
I think I originally made it ignore the `sequence_length` because seq2seq models always need variable-length inputs. Well, unless you're trying to work around Core ML limitations, I guess. ;-)
If you can share the mlmodel file with the error, I can take a look.
Run this script to fix the issue: ``` import coremltools import numpy as np mlmodel = coremltools.models.MLModel("bert-test-256_FP16.mlmodel") spec = mlmodel._spec spec.neuralNetwork.layers[9].activation.linear.alpha = 1 # whereNonZero spec.neuralNetwork.layers[11].activation.linear.alpha = 1 # squeeze...
It probably got inserted in the ONNX conversion step. There are most likely a whole bunch of other layers that don't really need to be in there. :-D
ONNX and Core ML do not have 100% the same features, so it's possible your ONNX model contains an operator that is not supported by Core ML. Sometimes you can...
You'll need to provide more info, @ITBackkom. Usually there is an additional error message that explains why CoreMLModelCompile failed (you can see this in the build log pane in Xcode)....
What is the file size of that model file when you look at it with Finder? This repo uses git-lfs to store the large model files. Most likely, your model...
That sounds like an interesting approach!
Honestly, it's been too long since I did anything with Core ML so I don't know off the top of my head.