Reason for Tensorflow on CPU evaluation?

[JonQKim]

Thank you for your sharing great works!!

In the README, you mentioned that the inference time on CPU was evaluated using Tensorflow. Would you share the reason why you tested on TF, even though you implemented your code using Pytorch? And would you also share how you tested on Tensorflow, like converting onnx to TF?

[cleardusk]

At that time, the inference speed of TF is faster than PyTorch. Now, ONNX is faster. I think there is no need to test the code on TF now?