Onnx to mlmodel conversion fails to generate .mlmodel file

1

I’m trying to convert Pytorch model to MLModel with Onnx.

My code:

import torch
from onnx_coreml import convert
import coremltools

net = BiSeNet(19)
net.cuda()
net.load_state_dict(torch.load('model.pth'))

dummy = torch.rand(1,3,512,512).cuda()

torch.onnx.export(net, dummy, "Model.onnx", input_names=["image"], output_names=["output"], opset_version=11)

finalModel = convert(model='Model.onnx', minimum_ios_deployment_target='12')
finalModel.save('ModelML.mlmodel')

After the code runs Model.onnx is generated, however, .mlmodel file is not generated. There’re no errors in the console.

What could be the issue?

2

Did you verify that the generated ONNX model is running fine?
If that’s the case, it seems that onnx_coreml fails to store it properly, so I would recommend to create an issue here.

3

How can I verify ONNX model? This is my model.

4

You could run it with e.g. onnxruntime as described here.