I have an input of 32×32×3 and I want to feed it to a convolution layer that output 32×32×16 feature maps. I know that I can compute a single feature map as matrix multiplication. But my question is, if it's possible to compute the whole feature maps at once.

Thank you in advance.

  • 1
    $\begingroup$ What do you mean by "at once" in "compute the whole feature maps at once"? At the same time or at the same place? Parallel computation? $\endgroup$ – Apass.Jack Jun 13 at 12:08

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.