Whydoweneed`Conv2dStaticSamePadding`

Sure, on it

Le mar. 23 juin 2020 17:46, Luke Melas-Kyriazi a
écrit :

I think you're right (at least now with all the advances in the recent
versions of PyTorch). I put in the static padding because right after this
repo was released, a bunch of people asked for exporting, and it was the
easiest way I knew of doing exporting at the time.

Do you think you might be able to write up a Colab notebook that does the
ONNX export with non-static padding? If it looks good, I can remove the
static padding and we can integrate your notebook into the README.

—
You are receiving this because you commented.
Reply to this email directly, view it on GitHub
https://github.com/lukemelas/EfficientNet-PyTorch/issues/122#issuecomment-648445018,
or unsubscribe
https://github.com/notifications/unsubscribe-auth/ABFLA3XOTCFYYOIFLT4W7J3RYEPDFANCNFSM4J6F4QHA
.