Jeshua Lacock

Ugh - created a new coremltools 4.1 with PyTorch 1.8.1 env and no longer work with flexible output shapes. Now I don't even know how to get the workaround working...

It seems as of 4.1 I flexible image output shapes no longer work and I can't get 4.0 installed because of a numpy float32 error. I was using 4.0 with...

Yes, my model works properly with PyTorch. It works if the input/output is not flexible with both images and arrays. I don't understand how that is not related to flexible...

I'll work on sharing reproducibles with you, but, I did not save scripts of everything I tried (and I tried countless configurations). So I have first go back and produce...

 Here is a full reproducible, model class, conversion script, PyTorch and MLModel: [demoModelColor.zip](https://github.com/apple/coremltools/files/6903848/demoModelColor.zip) Note that this is a slightly different architecture, but...

Sadly no.

Actually, using the skip_unpool model causes a core dump: ``` THCudaCheck FAIL file=/home/prime/torch/extra/cutorch/lib/THC/generic/THCStorage.c line=147 error=77 : an illegal memory access was encountered Segmentation fault (core dumped) ```

Hi @DmitryUlyanov would it be helpful for me to upload the models?

If we could get the trained model in Caffe or Keras then it should be possible. I tried converting a trained Torch .t7 model to a Caffe model using Facebook's...

You have to use the option "--bundle-resources-for-swift-cli" then specify that path for the "--resource-path".