Did you encounter an error from scale_factor in your Upsample layers (transformer_net.py, line 96) when using onnx_coreml.convert? It appears to be a known issue, but I'm wondering if you found a workaround. Thanks
@colinheye Sorry I didn't see this sooner! I don't recall, but I would have documented any workarounds. I have had issues with ONNX to CoreML but have filed issues on their GitHub for all of them.
Did you encounter an error from scale_factor in your Upsample layers (transformer_net.py, line 96) when using onnx_coreml.convert? It appears to be a known issue, but I'm wondering if you found a workaround. Thanks