Where is the 1*1 conv?

[xiaolongwu0713]

Hi, there.
Thanks for your code. But seems your implementation is different from the EEGnet paper. In the paper there is a depthwise conv lary which is a 1*1 conv. So I suppose this is not the same as the implementation in the paper?

[aliasvishnu]

EEGnet design has changed in 2018. I don't know of further revisions. This implement is of the first version. At that time, the authors hadn't released their code. Now they have. Please use the authors' original implementation linked in my repo itself.

…

On Sun, Jan 31, 2021, 3:21 PM xiaolongwu0713 ***@***.***> wrote: Hi, there. Thanks for your code. But seems your implementation is different from the EEGnet paper. In the paper there is a depthwise conv lary which is a 1*1 conv. So I suppose this is not the same as the implementation in the paper? — You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub <#4>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/ABFBDDLJ4IVXME2ICDG5XPTS4XQWLANCNFSM4W32GH7A> .