Why do we use ONNX to represent 'Open Neural Network Exchange' instead of ONNE?

[slowlyideal]

Ask a Question

Why do we use ONNX to represent 'Open Neural Network Exchange' instead of ONNE?

Question

I have been working with ONNX for a while, and recently I need to introduce ONNX to others. Hence, I have this simple question: Is it because of English grammar or is there another story behind it?

Further information

• Relevant Area (e.g. model usage, best practices, shape_inference, version_converter, training, test):

• Is this issue related to a specific model?
  Model name (e.g. mnist):
  Model opset (e.g. 7):
  Please upload the model and provide the link if possible.

Notes

Any additional information, code snippets.

[slowlyideal]

@jcwchen @xadupre anyone can help me?

[jcwchen]

Hi @slowlyideal, Interesting question. Sorry that I don't really have the history context here (haven't worked with ONNX when ONNX was born), but I would guess "X" is closer to the prefix of "exchange" (ex). Also possibly there is some branding concern like ONNE has been well used compared to ONNX.

[slowlyideal]

I am very grateful（hhhh, fei chang gan xie）

Hi @slowlyideal, Interesting question. Sorry that I don't really have the history context here (haven't worked with ONNX when ONNX was born), but I would guess "X" is closer to the prefix of "exchange" (ex). Also possibly there is some branding concern like ONNE has been well used compared to ONNX.

I am very grateful（hhhh, fei chang gan xie）

[jcwchen]

I will close it now. Feel free to reopen it if you still have other ONNX questions. Thanks!