I always know that kernel size should be smaller than input size. I wonder what if the opposite.

For example, if I have a 2d average pooling layer: Input size=4X4, Kernel size=6,Stride=1,Padding = SAME

I have coded this using tensorflow and prettytensor and it gives no error. But I wonder that is this a valid pooling? Does it mean something?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.