Toby Roseman

Sorry you're having this issue. Please give us a bit more information. Are you using macOS or Linux? Also please copy and paste the complete output. I want to see...

Hi @Ignasinou - Please post the complete output of your call to `tc.sound_classifier.create`. The sound classifier gets created [in three different stages](https://apple.github.io/turicreate/docs/userguide/sound_classifier/how-it-works.html). I want to see what stage is getting...

There is no good or easy way to do this. It's possible to get the spec from your `MLModel`. Then edit that protobuf to split that model into multiple models....

Core ML models are not really designed to be modified after they have been created. It's going to be easier to make the modification to your original model before you...

Thanks for the minimal example. Taking a look at `print(model)`, I see that the only input to the model (`my_in`) is a multiArray of FLOAT32 with shape 20. However `np.random.rand(20)`...

I’m seeing different behavior from numpy. What version of numpy are you using?

I can reproduce this issue with numpy `1.19.5` and coremltools `4.1` on macOS 11. However this problem **_seems_** to have already been fixed. If you upgrade to our latest pre-release...

Rebased change and kicked off a CI run: https://gitlab.com/coremltools1/coremltools/-/pipelines/630204718

Updated CI: https://gitlab.com/coremltools1/coremltools/-/pipelines/630212009

We just released coremltools 6.0. This version does support PyTorch 1.12.1. Are you converting to `neuralnetwork` or `mlprogram`? If you're converting to `neuralnetwork` or not specifying that value, try using...