Aseem Wadhwa

That means some part of the pytorch code is getting lowered into the `new_empty` op after tracing. Can you please check by printing `traced_model.forward.graph` which part of the code corresponds...

`NSLocalizedDescription = "Error binding image input buffer input_14.";` : is an unfortunately an error that occurs with the combination of "image inputs + rangeDim" on macOS Big Sur. The workaround...

@jral3s , can you please expand by what you mean by "model that is optimized for mobile"? Which torch APIs are you using in particular to "optimize" the model and...

The optimizations specified [here](https://pytorch.org/docs/stable/mobile_optimizer.html) are specific to pytorch mobile and not necessarily relevant or applicable or even supported for CoreML. The relevant CoreML optimizations are performed during conversion, and later...

> How can I file a feedback request for this model? https://developer.apple.com/bug-reporting/

- I don't think its good idea to pass additional flags via kwargs. If its something to be officially it should be part of the argument list - can you...

Yeah that would be great if you can add a test case (say in https://github.com/apple/coremltools/blob/master/coremltools/converters/mil/frontend/tensorflow2/test/test_v2_load.py or https://github.com/apple/coremltools/blob/master/coremltools/test/api/test_api_examples.py). I think actually, adding it as part of kwargs initially might be better,...

I just verified that enumerated shapes are only allowed for a single input. For multiple inputs, please use "range" flexibility instead. The error message should be improved. Also, the recommended...

I think this is probably not related to flexible output shape. It could also be an issue with the preview feature. Have you verified the Core ML model output by...

Can you please share the code that you are using to compare torch and coreml predictions, which passes when input/output are not flexible, but fails, when they are. Conversion code...